Trial Outcomes & Findings for Disseminating Public Health Evidence to Support Prevention and Control of Diabetes Among Local Health Departments (NCT NCT03211832)

NCT ID: NCT03211832

Last Updated: 2023-02-10

Results Overview

Survey participants were asked to rate the perceived importance of each of 10 skills pertinent to evidence-based decision making on an 11-point ordered scale (possible values 0 not at all important to 10 very important for each skill), and to rate the perceived availability in the agency of each of the same skills from 0-10 (0 not at all available to 10 very available). A skill gap was calculated for each skill as perceived importance minus perceived availability (possible values -10 to +10 for each skill). An overall EBDM skill gap was created by taking the average across all 10 skill gaps (possible values -10 to +10). Higher scores indicate a worse outcome. Skill items: community assessment; quantifying the issue; prioritization; action planning; adapting interventions; evaluation designs; quantitative evaluation; qualitative evaluation; economic evaluation; and communicating evidence to decision-makers. A definition for each was provided that started with the word "understand".

• Recruitment status: COMPLETED
• Study phase: NA
• Target enrollment: 331 participants
• Primary outcome timeframe: 24 months post baseline
• Results posted on: 2023-02-10

Participant Flow

12 local public health departments were recruited into the study in 3 groups. Once a health department agreed to participate, health department managers provided employee lists for self-report survey invitations.

Each group of 4 health departments crossed over into intervention at different times. In the stepped-wedge design, at baseline, surveys from all 12 health departments were analyzed as control surveys. Once a health department had crossed over into intervention, all remaining surveys of that health department's employees were analyzed as intervention surveys. At the last survey data collection, all surveys were analyzed as intervention surveys.

Unit of analysis: Health departments/agencies

Participant milestones

Measure	Group 1 - Control (2 Months), Then Intervention (24 Months) While in control status, the health departments will conduct usual public health practice. During Intervention, each health department will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.	Group 2 - Control (10 Months), Then Intervention (16 Months) While in control status, the health department conducted usual public health practice. Intervention began with a multi-day training in evidence-based decision making. Then each health department selected and helped develop intra-organizational procedures, policies, and additional activities to embed evidence-based decision making into day-to-day public health practice.	Group 3 - Control (18 Months), Then Intervention (8 Months) While in control status, the health department conducted usual public health practice. Intervention began with a multi-day training in evidence-based decision making. Then each health department selected and helped develop intra-organizational procedures, policies, and additional activities to embed evidence-based decision making into day-to-day public health practice.
Overall Study STARTED	78 4	85 4	66 4
Overall Study Survey Time 2	83 4	87 4	63 4
Overall Study Survey Time 3	70 4	112 4	55 4
Overall Study Survey Time 4	70 4	97 4	60 4
Overall Study COMPLETED	42 4	33 4	32 4
Overall Study NOT COMPLETED	36 0	52 0	34 0

Reasons for withdrawal

Measure	Group 1 - Control (2 Months), Then Intervention (24 Months) While in control status, the health departments will conduct usual public health practice. During Intervention, each health department will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.	Group 2 - Control (10 Months), Then Intervention (16 Months) While in control status, the health department conducted usual public health practice. Intervention began with a multi-day training in evidence-based decision making. Then each health department selected and helped develop intra-organizational procedures, policies, and additional activities to embed evidence-based decision making into day-to-day public health practice.	Group 3 - Control (18 Months), Then Intervention (8 Months) While in control status, the health department conducted usual public health practice. Intervention began with a multi-day training in evidence-based decision making. Then each health department selected and helped develop intra-organizational procedures, policies, and additional activities to embed evidence-based decision making into day-to-day public health practice.
Overall Study No longer with agency	32	49	30
Overall Study Declined survey	3	3	2
Overall Study Not available during data collection	1	0	2

Baseline Characteristics

Race/Ethnicity was not collected, therefore the number analyzed of zero is different from the overall number of surveys analyzed for the other measures.

Baseline characteristics by cohort

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.	Total n=922 completed surveys Total of all reporting groups
Age, Customized 18-29 years	103 completed surveys n=433 completed surveys	92 completed surveys n=489 completed surveys	195 completed surveys n=922 completed surveys
Age, Customized 30-39 years	111 completed surveys n=433 completed surveys	127 completed surveys n=489 completed surveys	238 completed surveys n=922 completed surveys
Age, Customized 40-49 years	88 completed surveys n=433 completed surveys	100 completed surveys n=489 completed surveys	188 completed surveys n=922 completed surveys
Age, Customized 50-59 years	88 completed surveys n=433 completed surveys	116 completed surveys n=489 completed surveys	204 completed surveys n=922 completed surveys
Age, Customized >=60 years	42 completed surveys n=433 completed surveys	52 completed surveys n=489 completed surveys	94 completed surveys n=922 completed surveys
Age, Customized Not reported	1 completed surveys n=433 completed surveys	2 completed surveys n=489 completed surveys	3 completed surveys n=922 completed surveys
Sex: Female, Male Female	356 completed surveys n=433 completed surveys	411 completed surveys n=489 completed surveys	767 completed surveys n=922 completed surveys
Sex: Female, Male Male	77 completed surveys n=433 completed surveys	78 completed surveys n=489 completed surveys	155 completed surveys n=922 completed surveys
Race/Ethnicity, Customized	—	—	0 completed surveys Race/Ethnicity was not collected, therefore the number analyzed of zero is different from the overall number of surveys analyzed for the other measures.
Region of Enrollment United States	433 completed surveys n=433 completed surveys	489 completed surveys n=489 completed surveys	922 completed surveys n=922 completed surveys

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Survey participants were asked to rate the perceived importance of each of 10 skills pertinent to evidence-based decision making on an 11-point ordered scale (possible values 0 not at all important to 10 very important for each skill), and to rate the perceived availability in the agency of each of the same skills from 0-10 (0 not at all available to 10 very available). A skill gap was calculated for each skill as perceived importance minus perceived availability (possible values -10 to +10 for each skill). An overall EBDM skill gap was created by taking the average across all 10 skill gaps (possible values -10 to +10). Higher scores indicate a worse outcome.

Skill items: community assessment; quantifying the issue; prioritization; action planning; adapting interventions; evaluation designs; quantitative evaluation; qualitative evaluation; economic evaluation; and communicating evidence to decision-makers. A definition for each was provided that started with the word "understand".

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Evidence-based Decision Making (EBDM) Competencies	2.05 score on a scale Interval 1.87 to 2.23	1.98 score on a scale Interval 1.82 to 2.13

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Self-reported number of evidence-based chronic disease prevention interventions implemented by the local health department from a pre-populated list of 8 evidence-based interventions to prevent diabetes and other chronic diseases (possible score 0 to 8). Higher score indicates a better outcome.

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Evidence-based Intervention Score	4.84 units on a scale Interval 4.61 to 5.07	4.58 units on a scale Interval 4.36 to 4.8

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Self-report Likert scale items measure personal awareness of opportunities to learn about and apply EBDM among local level chronic disease control public health practitioners. Each item is measured on a 7-point Likert scale from 1 strongly disagree to 7 strongly agree. A summary score was created as an average of the items within the domain. Possible scores 1 to 7. Higher scores mean a better outcome.

The items were: a) I am provided the time to identify evidence-based programs and practices; b) My direct supervisor recognizes the value of management practices that facilitate evidence-based decision-making; c) My work group/division offers employees opportunities to attend evidence-based-decision making trainings; and d) Top leadership in my agency (e.g., director, assistant directors) recognizes the value of evidence-based decision-making.

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Awareness of Culture Supportive of EBDM	5.34 score on a scale Interval 5.22 to 5.47	5.43 score on a scale Interval 5.31 to 5.54

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Self-report Likert scale items assess perceived supervisory expectations for EBDM use and perceived work unit/division capacity to carry out EBDM. Each item is measured on a 7-point Likert scale from 1 strongly disagree to 7 strongly agree. A summary score was created as an average of the items (possible scores 1 to 7). Higher scores mean a better outcome.

Items: a) I use EBDMin my work; b) My direct supervisor expects me to use evidence-based decision making; c) My performance is partially evaluated on how well I use evidence-based decision making in my work; d) My work group/division currently has the resources (e.g. staff, facilities, partners) to support application of evidence-based decision making; e) The staff in my work group/division has the necessary skills to carry out evidence-based decision making; f) The majority of my work group/division's external partners support use of EBDM; and g) Top leadership in my agency encourages use of EBDM.

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Capacity and Expectations for Evidence-based Decision Making (EBDM)	5.22 score on a scale Interval 5.11 to 5.33	5.24 score on a scale Interval 5.14 to 5.35

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Self-report Likert scale items measured perceived work unit's resource availability for evidence-based decision making. Each item is measured on a 7-point Likert scale from 1 strongly disagree to 7 strongly agree. A summary score was created as an average of the items. Possible scores 1 to 7. Higher scores mean a better outcome.

The items in the scale were: a) Informational resources (e.g. academic journals, guidelines, and toolkits) are available to my work group/division to promote the use of evidence-based decision making; b) My work group/division engages a diverse external network of partners that share resources to facilitate evidence-based decision making; and c) Stable funding is available for evidence-based decision making.

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Resource Availability	4.46 score on a scale Interval 4.32 to 4.6	4.47 score on a scale Interval 4.35 to 4.6

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Self-report Likert scale of work unit's support of community needs assessment, utilization of evaluation for pre and post program implementation as well as for dissemination purposes. Each item is measured on a 7-point Likert scale from 1 strongly disagree to 7 strongly agree. A summary score was created as an average of the items within the domain. Possible scores 1 to 7. Higher scores mean a better outcome.

Items: a) My work group/division supports community needs assessments to ensure that evidence-based decision-making approaches continue to meet community needs; b) My work group/division plans for evaluation of interventions prior to implementation; c) My work group/division uses evaluation data to monitor and improve interventions; and d) My work group/division distributes intervention evaluation findings to other organizations that can use our findings.

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Evaluation Capacity of Work Unit	5.23 score on a scale Interval 5.1 to 5.33	5.19 score on a scale Interval 5.07 to 5.31

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Self-report Likert scale assessing perceived health department culture supportive of EBDM, information sharing and participatory decision making. Each item is measured on a 7-point Likert scale from 1 strongly disagree to 7 strongly agree. A summary score was created as an average of the items within the domain. Possible scores 1 to 7. Higher scores mean a better outcome.

Items: a) My work group/division has access to evidence-based decision making information that is relevant to community needs; b) When decisions are made within my work group/division, program staff members are asked for input; c) Information is widely shared in my work group/division so that everyone who makes decisions has access to all available knowledge; d) My agency is committed to hiring people with relevant training or experience in public health core disciplines (e.g., epidemiology, health education, environmental health); and e) My agency has a culture that supports the processes necessary for EBDM.

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
EBDM Climate Cultivation	5.21 score on a scale Interval 5.08 to 5.34	5.26 score on a scale Interval 5.14 to 5.37

PRIMARY outcome

Timeframe: 24 months post baseline

Population: The unit of analysis is completed surveys due to the stepped-wedge design and having a different set of survey participants at each of the 4 survey time points. All baseline surveys were analyzed as control; at T4 all surveys were analyzed as intervention. Survey participants were health department employees. Due to staff turnover, some had left the agency between each data collection period and others hired, so additional health department staff were invited at Survey T2, T3, and T4.

Self-report Likert scale items assess perceived importance of partnering across sectors to share resources and address population health issues. Each item is measured on a 7-point Likert scale from 1 strongly disagree to 7 strongly agree. A summary score was created as an average of the items within the domain. Possible scores 1 to 7. Higher scores mean a better outcome.

Items: a) Our collaborative partnerships have missions that align with my agency; b) It is important to my agency to have partners who share resources (money, staff time, space, materials); c) It is important to my agency to have partners in health care to address population health issues; and d) It is important to my agency to have partners in other sectors (outside of health) to address population health issues

Outcome measures

Measure	Control n=433 completed surveys The control group will conduct usual public health practice.	Intervention n=489 completed surveys Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Partnerships to Support EBDM	5.94 score on a scale Interval 5.86 to 6.03	5.91 score on a scale Interval 5.83 to 6.0

SECONDARY outcome

Timeframe: 24 months post baseline

Population: Agencies with completed social network surveys at baseline when all health departments were in control status and at the end of the study when all health departments were in the intervention.

The average number of links per agency is the measure of connectedness with other agencies that is reported here. The measure is from a separate self-report social network survey.

Outcome measures

Measure	Control n=224 Agencies The control group will conduct usual public health practice.	Intervention n=261 Agencies Participating local health departments will help develop and choose several dissemination activities they prefer for their local health department to receive. Dissemination activities may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes and chronic disease prevention and control. Dissemination of public health knowledge: Participating local health departments will help develop and choose dissemination strategies they prefer for their staff working in and supporting diabetes and chronic disease prevention and control to receive. Dissemination strategies may include multi-day in-person training workshops, electronic information exchange modalities, remote technical assistance, and information on ways to enhance organizational climates favorable to evidence-based diabetes prevention and control.
Inter-agency Connectedness	6.8 Agencies Standard Deviation 1.7	8.5 Agencies Standard Deviation 2.7

Adverse Events

Control

Serious events: 0 serious events

Other events: 0 other events

Deaths: 0 deaths

Intervention

Serious events: 0 serious events

Other events: 0 other events

Deaths: 0 deaths

Serious adverse events

Adverse event data not reported

Other adverse events

Adverse event data not reported

Additional Information

Ross Brownson

Prevention Research Center, Brown School, Washington University in St. Louis

Phone: 3149350121

Email: rbrownson@wustl.edu

Results disclosure agreements

• Principal investigator is a sponsor employee
• Publication restrictions are in place