Dataset Information

A Safe and Practical Cycle for Team-Based Development and Implementation of In-House Clinical Software.

ABSTRACT:

Purpose

Due to a gap in published guidance, we describe our robust cycle of in-house clinical software development and implementation, which has been used for years to facilitate the safe treatment of all patients in our clinics.

Methods and materials

Our software development and implementation cycle requires clarity in communication, clearly defined roles, thorough commissioning, and regular feedback. Cycle phases include design requirements and use cases, development, physics evaluation testing, clinical evaluation testing, and full clinical release. Software requirements, release notes, test suites, and a commissioning report are created and independently reviewed before clinical use. Software deemed to be high-risk, such as those that are writable to a database, incorporate the use of a formal, team-based hazard analysis. Incident learning is used to both guide initial development and improvements as well as to monitor the safe use of the software.

Results

Our standard process builds in transparency and establishes high expectations in the development and use of custom software to support patient care. Since moving to a commercial planning system platform in 2013, we have applied our team-based software release process to 16 programs related to scripting in the treatment planning system for the clinic.

Conclusions

The principles and methodology described here can be implemented in a range of practice settings regardless of whether or not dedicated resources are available for software development. In addition to teamwork with defined roles, documentation, and use of incident learning, we strongly recommend having a written policy on the process, using phased testing, and incorporating independent oversight and approval before use for patient care. This rigorous process ensures continuous monitoring for and mitigatation of any high risk hazards.

SUBMITTER: Moran JM

PROVIDER: S-EPMC8767245 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundThis study aimed to (1) develop a fully residual deep convolutional neural network (CNN)-based segmentation software for computed tomography image segmentation of the male pelvic region and (2) demonstrate its efficiency in the male pelvic region.MethodsA total of 470 prostate cancer patients who had undergone intensity-modulated radiotherapy or volumetric-modulated arc therapy were enrolled. Our model was based on FusionNet, a fully residual deep CNN developed to semantically segment biological images. To develop the CNN-based segmentation software, 450 patients were randomly selected and separated into the training, validation and testing groups (270, 90, and 90 patients, respectively). In Experiment 1, to determine the optimal model, we first assessed the segmentation accuracy according to the size of the training dataset (90, 180, and 270 patients). In Experiment 2, the effect of varying the number of training labels on segmentation accuracy was evaluated. After determining the optimal model, in Experiment 3, the developed software was used on the remaining 20 datasets to assess the segmentation accuracy. The volumetric dice similarity coefficient (DSC) and the 95th-percentile Hausdorff distance (95%HD) were calculated to evaluate the segmentation accuracy for each organ in Experiment 3.ResultsIn Experiment 1, the median DSC for the prostate were 0.61 for dataset 1 (90 patients), 0.86 for dataset 2 (180 patients), and 0.86 for dataset 3 (270 patients), respectively. The median DSCs for all the organs increased significantly when the number of training cases increased from 90 to 180 but did not improve upon further increase from 180 to 270. The number of labels applied during training had a little effect on the DSCs in Experiment 2. The optimal model was built by 270 patients and four organs. In Experiment 3, the median of the DSC and the 95%HD values were 0.82 and 3.23 mm for prostate; 0.71 and 3.82 mm for seminal vesicles; 0.89 and 2.65 mm for the rectum; 0.95 and 4.18 mm for the bladder, respectively.ConclusionsWe have developed a CNN-based segmentation software for the male pelvic region and demonstrated that the CNN-based segmentation software is efficient for the male pelvic region.

Project description:BackgroundThis study was a retrospective evaluation of an unsuccessfully implemented team-based, chronic disease management program, with an aim to understand more about implementation barriers. The program, the Chronic Disease Management Initiative (CDMI) was a new collaborative model of care for patients with COPD. It utilized customized health information and interactive tools, mainly smartphones, for ongoing disease management. The program's goal was to demonstrate that integrated team-based models of care could improve patient care, as well as reduce readmission rates and emergency department visits. The program planning for CDMI began in 2017, followed by the implementation and evaluation period in 2018. After a 10-month implementation period, the program was unable to enroll a sufficient number of patients to examine if there was an improvement in patient outcomes.MethodsA retrospective case-study design using multiple data sources was used to gather feedback from participants involved in CDMI. Data collection occurred throughout planning and implementation and continued into early 2019. Semi-structured interviews were conducted, and transcripts were analyzed using NVivo 10 software. This was followed by content analysis.ResultsAnalysis revealed four key themes as barriers to CDMI's implementation: 1) lack of a needs assessment with key stakeholders; 2) lack of buy-in from medical staff; 3) inadequate patient engagement and; 4) contextual barriers. Planners did not conduct a proper needs assessment, nor include patients in the study design. In addition, there was insufficient consideration for how CDMI should be integrated into the usual COPD care plan, leading to confusion in roles and responsibilities. Poor communication between the implementation team and healthcare providers implementing the program, led to a lack of buy-in and engagement.ConclusionThe key themes resonate with what is already known in the literature. This study supports the importance of using a theoretically grounded plan for implementation. Using a model only in the planning stages is insufficient even when an intervention is based on evidence to support higher quality care. It is imperative to consider practical and contextual factors of program implementation and their interactions. By detailing the 'failed implementation' of this intervention, we hope to share important lessons about the need to plan implementation processes early in program planning.

Project description:BackgroundCommunication within the circle of care is central to coordinated, safe, and effective care; yet patients, caregivers, and health care providers often experience poor communication and fragmented care. Through a sequential program of research, the Loop Research Collaborative developed a web-based, asynchronous clinical communication system for team-based care. Loop assembles the circle of care centered on a patient, in private networking spaces called Patient Loops. The patient, their caregiver, or both are part of the Patient Loop. The communication is threaded, it can be filtered and sorted in multiple ways, it is securely stored, and can be exported for upload to a medical record.ObjectiveThe objective of this study was to implement and evaluate Loop. The study reporting adheres to the Standards for Reporting Implementation Research.MethodsThe study was a hybrid type II mixed methods design to simultaneously evaluate Loop's clinical and implementation effectiveness, and implementation barriers and facilitators in 6 health care sites. Data included monthly user check-in interviews and bimonthly surveys to capture patient or caregiver experience of continuity of care, in-depth interviews to explore barriers and facilitators based on the Consolidated Framework for Implementation Research (CFIR), and Loop usage extracted directly from the Loop system.ResultsWe recruited 25 initiating health care providers across 6 sites who then identified patients or caregivers for recruitment. Of 147 patient or caregiver participants who were assessed and met screening criteria, 57 consented and 52 were enrolled on Loop, creating 52 Patient Loops. Across all Patient Loops, 96 additional health care providers consented to join the Loop teams. Loop usage was followed for up to 8 months. The median number of messages exchanged per team was 1 (range 0-28). The monthly check-in and CFIR interviews showed that although participants acknowledged that Loop could potentially fill a gap, existing modes of communication, workflows, incentives, and the lack of integration with the hospital electronic medical records and patient portals were barriers to its adoption. While participants acknowledged Loop's potential value for engaging the patient and caregiver, and for improving communication within the patient's circle of care, Loop's relative advantage was not realized during the study and there was insufficient tension for change. Missing data limited the analysis of continuity of care.ConclusionsFundamental structural and implementation challenges persist toward realizing Loop's potential as a shared system of asynchronous communication. Barriers include health information system integration; system, organizational, and individual tension for change; and a fee structure for health care provider compensation for asynchronous communication.