Executive Summary of the Assessing Surgical Site Infection Surveillance Technologies (ASSIST) Project

Background: The expert panel that conducted the Assessing Surgical Site Infection Surveillance Technologies (ASSIST) project elaborates on the key findings of the health technologies assessment (HTA) report in a series of articles addressing topics from workflow challenges to implementation strategies to new big data analytics tailored to incorporate serial patient-generated health data (PGHD). Conclusion: By reporting on the methodology, with an emphasis on stakeholder engagement, the ASSIST investigators provide the basis for a future deep dive into the next phase of PGHD integration into surgical site infection (SSI) surveillance.

A Roadmap for Automatic Surgical Site Infection Detection and Evaluation Using User-Generated Incision Images

Background: Emerging technologies such as smartphones and wearable sensors have enabled the paradigm shift to new patient-centered healthcare, together with recent mobile health (mHealth) app development. One such promising healthcare app is incision monitoring based on patient-taken incision images. In this review, challenges and potential solution strategies are investigated for surgical site infection (SSI) detection and evaluation using surgical site images taken at home. Methods: Potential image quality issues, feature extraction, and surgical site image analysis challenges are discussed. Recent image analysis and machine learning solutions are reviewed to extract meaningful representations as image markers for incision monitoring. Discussions on opportunities and challenges of applying these methods to derive accurate SSI prediction are provided. Conclusions: Interactive image acquisition as well as customized image analysis and machine learning methods for SSI monitoring will play critical roles in developing sustainable mHealth apps to achieve the expected outcomes of patient-taken incision images for effective out-of-clinic patient-centered healthcare with substantially reduced cost.

Identification of Important Features in Mobile Health Applications for Surgical Site Infection Surveillance

Background: A landscape analysis of mobile health (mHealth) applications and published literature related to their use in surgical site infection (SSI) detection and surveillance was conducted by the Assessing Surgical Site Infection Surveillance Technologies (ASSIST) investigators. Methods: The literature review focused on post-discharge SSI detection or tracking by caregivers or patients using mHealth technology. This report is unique in its review across both commercial and research-based mHealth apps. Apps designed for long-term wound tracking and those focused on care coordination and scheduling were excluded. A structured evaluation framework was used to assess the operational, technical, and policy features of the apps. Results: Of the 10 apps evaluated, only two were in full clinical use. A variety of data were captured by the apps including wound photographs (eight apps), wound measurements (three apps), dressing assessments (two apps), physical activity metrics (three apps), medication adherence (three apps) as well as structured surveys, signs, and symptoms. Free-text responses were permitted by at least two apps. The extent of integration with the native electronic health record system was variable. Conclusion: The examination of rapidly evolving technologies is challenged by lack of standard evaluative methods, such as those more commonly used in clinical research. This review is unique in its application of a structured evaluation framework across both commercial and research-based mHealth apps.

Technological Advances in Clinical Definition and Surveillance Methodology for Surgical Site Infection Incorporating Surgical Site Imaging and Patient-Generated Health Data

Background: Surgical site infection (SSI) continues to be a common and costly complication after surgery. The current commonly used definitions of SSI were devised more than two decades ago and do not take in to account more modern technology that could be used to make diagnosis more consistent and precise. Patient-generated health data (PGHD), including digital imaging, may be able to fulfill this objective. Methods: The published literature was examined to determine the current state of development in terms of using digital imaging as an aide to diagnose SSI. This information was used to devise possible methodology that could be used to integrate digital images to more objectively define SSI, as well as using these data for both surveillance activities and clinical management. Results: Digital imaging is a highly promising means to help define and diagnose SSI, particularly in remote settings. Multiple groups continue to actively study these emerging technologies, however, present methods remain based generally on subjective rather than objective observations. Although current images may be useful on a case-by-case basis, similar to physical examination information, integrating imaging in the definition of SSI to allow more automated diagnosis in the future will require complex image analysis combined with other available quantified data. Conclusions: Digital imaging technology, once adequately evolved, should become a cornerstone of the criteria for both the clinical and surveillance definitions of SSI.

Engaging Patients in Co-Design of Mobile Health Tools for Surgical Site Infection Surveillance: Implications for Research and Implementation

Background: As the use of patient-owned devices, including smartphones and tablets, to manage day-to-day activities grows, so does healthcare industry's interest to better leverage technology to engage patients. For surgical care, a unique opportunity exists to capture patient-generated health data (PGHD) including photographs. As part of a broader initiative to evaluate PGHD for surgical site infection (SSI) surveillance, we sought evidence regarding patient involvement and experience with PGHD for SSI monitoring and surveillance. Methods: Through a scoping review of the literature and semi-structured stakeholder interviews we gathered evidence on what is currently known about patient perspectives of and experiences with mobile health (mHealth) interventions for post-operative recovery. We presented findings to and discussed with the ASSIST PGHD Stakeholder Advisory Group (PSAG) to generate priorities for further examination. Results: Our scoping review yielded 34 studies that addressed post-discharge use of PGHD for monitoring and surveillance of SSI. Of these, 16 studies addressed at least one outcome regarding patient experience; the most commonly measured outcome was patient satisfaction. Only three studies reported on patient involvement in the development of PGHD tools and interventions. We conducted interviews (n = 24) representing a range of stakeholder perspectives. Interviewees stressed the importance of patient involvement in tool and program design, noting patient involvement ensures the "work" that patients do in their daily lives to manage their health and healthcare is recognized. Discussion of evidence with the ASSIST PSAG resulted in formal recommendations for direct involvement of patients and caregivers for future work. Conclusions: While mHealth initiatives to advance post-operative management offer the ability to improve patient engagement, work is needed to ensure the patient voice is reflected. Active engagement with patients and caregivers in the development of new technology, the design of new workflows, and the conduct of research and evaluation ensures that the patient experiences and values are incorporated.

Patient-Generated Health Data in Surgical Site Infection: Changing Clinical Workflow and Care Delivery

Background: The patient's history of present illness provides an important part of the data with which clinicians diagnose and treat. Once surgical patients are discharged, the ability to incorporate direct observation requires coordinating patient and provider for a clinical visit. Mobile technologies offer the ability to gather and organize the patient's history, supplement that history with photographs and other clinical observations, and convey those data accurately and rapidly to the entire clinical team. Methods: We review our experience with patient-generated health data in surgical site infection, draw parallels with similar work in other domains, and identify principles we have found useful. Results: Health information system implementations require substantial changes in provider workflow. Shared expectations between the patient and the surgical team, an incremental approach to change in clinical processes, and an emphasis on clinical utility all support successful implementation. Conclusions: The data collection and rapid information exchange afforded by monitoring post-operative, post-discharge patients using mobile technologies can support the expectations of both patients and providers and may provide a novel data source for public health surveillance of surgical site infection. Both uses of these data require careful attention to introducing changes in clinical workflow.

A Stakeholder-Driven Framework for Evaluating Surgical Site Infection Surveillance Technologies

Background: Patients increasingly use mobile devices to send text messages and photographic data to surgeons. There is potential to harness this patient-generated health data (PGHD) for clinical and public health surveillance of surgical site infection (SSI). Leveraging PGHD collected via remote monitoring in the post-operative period has the potential to produce important benefits for patients, surgeons, care teams, and infection surveillance and prevention. Methods: We conducted a health technology assessment (HTA), drawing heavily on stakeholder engagement to better understand current and potential uses of PGHD in post-operative care. Stakeholder engagement activities included assembling an advisory board composed of stakeholder experts, interviewing key informants, and seeking out stakeholder guidance to synthesize evidence from interviews, literature review, and technical app review in order to develop recommendations on the use of PGHD in SSI surveillance. Results: We conducted a review of the published literature, a technical/market scan of available apps for capturing post-operative PGHD, and two rounds of key informant interviews with stakeholders. In addition, we held a day-long workshop to solicit stakeholder feedback on initial findings of the project and to guide additional work. These activities culminated in an HTA report that provides guidance and recommendations on the use of PGHD in SSI surveillance, including practice, research, and public health surveillance, and identifies open issues on post-operative use of PGHD for which additional evidence and experience are needed to optimize application of those data for clinical and public health purposes. Conclusion: Stakeholders, individuals with direct experience, or interest in a given topic are critical to the HTA process. They provide insight to guide the work conducted, ensure that the topics addressed are relevant and important, and that products of the work are accessible and meaningful to the individuals who will be most impacted.

Impact of Intravenous Immunoglobulin on Survival in Necrotizing Fasciitis With Vasopressor-Dependent Shock: A Propensity Score-Matched Analysis From 130 US Hospitals

Background

Shock frequently complicates necrotizing fasciitis (NF) caused by group A Streptococcus (GAS) or Staphylococcus aureus. Intravenous immunoglobulin (IVIG) is sometimes administered for presumptive toxic shock syndrome (TSS), but its frequency of use and efficacy are unclear.

Methods

Adult patients with NF and vasopressor-dependent shock undergoing surgical debridement from 2010 to 2014 were identified at 130 US hospitals. IVIG cases were propensity-matched and risk-adjusted. The primary outcome was in-hospital mortality and the secondary outcome was median length of stay (LOS).

Results

Of 4127 cases of debrided NF with shock at 121 centers, only 164 patients (4%) at 61 centers received IVIG. IVIG subjects were younger with lower comorbidity indices, but higher illness severity. Clindamycin and vasopressor intensity were higher among IVIG cases, as was coding for TSS and GAS. In-hospital mortality did not differ between matched IVIG and non-IVIG groups (crude mortality, 27.3% vs 23.6%; adjusted odds ratio, 1.00 [95% confidence interval, .55-1.83]; P = .99). Early IVIG (≤2 days) did not alter this effect (P = .99). Among patients coded for TSS, GAS, and/or S. aureus, IVIG use was still unusual (59/868 [6.8%]) and lacked benefit (P = .63). Median LOS was similar between IVIG and non-IVIG groups (26 [13-49] vs 26 [11-43]; P = .84). Positive predictive values for identifying true NF and debridement among IVIG cases using our algorithms were 97% and 89%, respectively, based on records review at 4 hospitals.

Conclusions

Adjunctive IVIG was administered infrequently in NF with shock and had no apparent impact on mortality or hospital LOS beyond that achieved with debridement and antibiotics.

Comparison of Outcomes of antibiotic Drugs and Appendectomy (CODA) trial: a protocol for the pragmatic randomised study of appendicitis treatment

INTRODUCTION

Several European studies suggest that some patients with appendicitis can be treated safely with antibiotics. A portion of patients eventually undergo appendectomy within a year, with 10%-15% failing to respond in the initial period and a similar additional proportion with suspected recurrent episodes requiring appendectomy. Nearly all patients with appendicitis in the USA are still treated with surgery. A rigorous comparative effectiveness trial in the USA that is sufficiently large and pragmatic to incorporate usual variations in care and measures the patient experience is needed to determine whether antibiotics are as good as appendectomy.

OBJECTIVES

The Comparing Outcomes of Antibiotic Drugs and Appendectomy (CODA) trial for acute appendicitis aims to determine whether the antibiotic treatment strategy is non-inferior to appendectomy.

METHODS/ANALYSIS

CODA is a randomised, pragmatic non-inferiority trial that aims to recruit 1552 English-speaking and Spanish-speaking adults with imaging-confirmed appendicitis. Participants are randomised to appendectomy or 10 days of antibiotics (including an option for complete outpatient therapy). A total of 500 patients who decline randomisation but consent to follow-up will be included in a parallel observational cohort. The primary analytic outcome is quality of life (measured by the EuroQol five dimension index) at 4 weeks. Clinical adverse events, rate of eventual appendectomy, decisional regret, return to work/school, work productivity and healthcare utilisation will be compared. Planned exploratory analyses will identify subpopulations that may have a differential risk of eventual appendectomy in the antibiotic treatment arm.

ETHICS AND DISSEMINATION

This trial was approved by the University of Washington's Human Subjects Division. Results from this trial will be presented in international conferences and published in peer-reviewed journals.

TRIAL REGISTRATION NUMBER

NCT02800785.

Longer-Duration Antimicrobial Therapy Does Not Prevent Treatment Failure in High-Risk Patients with Complicated Intra-Abdominal Infections

BACKGROUND

Recent studies have suggested the length of treatment of intra-abdominal infections (IAIs) can be shortened without detrimental effects on patient outcomes. However, data from high-risk patient populations are lacking. We hypothesized that patients at high risk for treatment failure will benefit from a longer course of antimicrobial therapy.

METHODS

Patients enrolled in the Study to Optimize Peritoneal Infection Therapy (STOP-IT) trial were evaluated retrospectively to identify risk factors associated with treatment failure, which was defined as the composite outcome of recurrent IAI, surgical site infection, or death. Variables were considered risk factors if there was a positive statistical association with treatment failure. Patients were then stratified according to the presence and number of these risk factors. Univariable analyses were performed using the Kruskal-Wallis, χ², and Fisher exact tests. Logistic regression controlling for risk factors and original randomization group, either a fixed four-day antimicrobial regimen (experimental) or a longer course based on clinical response (control), also was performed.

RESULTS

We identified corticosteroid use, Acute Physiology and Chronic Health Evaluation II score ≥5, hospital-acquired infection, or a colonic source of IAI as risk factors associated with treatment failure. Of the 517 patients enrolled, 263 (50.9%) had one or two risk factors and 16 (3.1%) had three or four risk factors. The rate of treatment failure rose as the number of risk factors increased. When controlling for randomization group, the presence and number of risk factors were independently associated with treatment failure, but the duration of antimicrobial therapy was not.

CONCLUSIONS

We were able to identify patients at high risk for treatment failure in the STOP-IT trial. Such patients did not benefit from a longer course of antibiotic administration. Further study is needed to determine the optimum duration of antimicrobial therapy in high-risk patients.

Wound Concerns and Healthcare Consumption of Resources after Colorectal Surgery: An Opportunity for Innovation?

BACKGROUND

Significant portions of patients undergoing colorectal surgical procedures have minor incision disturbances, yet very few meet definitions for surgical site infection (SSI). We sought to investigate the natural history of incision disturbances with a focus on the patient experience and resource utilization. We hypothesize that patients who have an incision disturbance consume frequent healthcare resources in the post-operative period despite the fact that most never receive a diagnosis of SSI.

METHODS

A 24-month prospective observational study was undertaken at an academic institution. Patients undergoing elective colorectal operation by two board-certified colorectal surgeons were followed prospectively for 90 days. Incisions were photographed serially and clinically characterized beginning as early as post-operative day two and at follow-up visits. The primary outcome was patient concern for an incision disturbance. Three surgeons reviewed clinical data and photographs to determine the presence of an incisional surgical infection, and diagnosis required agreement from two of three surgeons.

RESULTS

There were 171 patients included; 31 (15%) sought evaluation from a healthcare provider for concerns related to their incision including 46 telephone calls, six emergency department visits, seven primary care visits, 10 home health and 40 surgical clinic visits. Incision erythema and drainage were the most common sources of patient concern. Mean body mass index was higher in patients with concern for incision disturbances (34 vs. 28 kg/m², p < 0.0001). Ultimately, 8% (14/171) received a diagnosis of SSI by study criteria while only 2% (4/171) were captured as having an SSI by the institutional National Surgical Quality Improvement Program database (p < 0.0001).

CONCLUSIONS

Patients undergoing colorectal surgical procedures commonly are concerned with post-operative incision disturbance, yet few are associated with a diagnosis of SSI, and in-person evaluation yields frequent utilization of healthcare resources. This presents an opportunity for secure electronic communication with the surgical team and the patient to potentially reduce consumption of healthcare resources.

Outcomes in necrotizing soft tissue infections treated with therapeutic plasma exchange

BACKGROUND

Treatment of necrotizing soft tissue infections (NSTIs) includes prompt surgical debridement and antibiotics, but despite standard care, the morbidity and mortality remain high. Since therapeutic plasma exchange (TPE) has been considered for treatment of severe sepsis, this study evaluates the efficacy of TPE for patients with NSTI.

STUDY DESIGN AND METHODS

This is a retrospective study of patients with diagnosis of NSTI who received treatment with and without TPE over an 11-year period. The primary outcome was in-hospital mortality.

RESULTS

Fifty-two patients with NSTI treated with TPE (TPE group) and 125 patients with NSTI not treated with TPE (non-TPE group) were assessed. Nineteen (36.5%) patients died in the TPE group, and 35 (28%) patients died in the non-TPE group. Within the TPE group, there was significant improvement in white blood cell (WBC) count and sodium levels 7 days after TPE treatment, but no improvement in creatinine. Inverse probability weighting based on propensity scores was used to compare survival in the TPE and non-TPE groups and demonstrated that TPE was associated with an increased odds of death (odds ratio, 2.8). A second analysis matched for six variables yielded 31 pairs and demonstrated no significant difference in mortality or length of stay.

CONCLUSIONS

This study describes the largest series of patients with NSTIs treated with TPE and showed no evidence of clinical benefit. Further carefully designed studies with meaningful clinical endpoints would prove useful in assessing reproducibility and determining if there is a role for TPE in other forms of severe sepsis.

Diagnosing Surgical Site Infection Using Wound Photography: A Scenario-Based Study

BACKGROUND

Postoperative surgical site infections (SSI) are common and costly. Most occur post discharge, and can result in potentially preventable readmission or unnecessary urgent evaluation. Mobile health approaches incorporating patient-generated wound photos are being implemented in an attempt to optimize triage and management. We assessed how adding wound photos to existing data sources modifies provider decision making.

STUDY DESIGN

We used a web-based simulation survey using a convenience sample of providers with expertise in surgical infections. Participants viewed a range of scenarios, including surgical history, physical exam, and description of wound appearance. All participants reported SSI diagnosis, diagnostic confidence, and management recommendations (main outcomes) first without, and then with, accompanying wound photos. At each step, participants ranked the most important features contributing to their decision.

RESULTS

Eighty-three participants completed a median of 5 scenarios (interquartile range 4 to 7). Most participants were physicians in academic surgical specialties (n = 70 [84%]). The addition of photos improved overall diagnostic accuracy from 67% to 76% (p < 0.001), and increased specificity from 77% to 92% (p < 0.001), but did not significantly increase sensitivity (55% to 65%; p = 0.16). Photos increased mean confidence in diagnosis from 5.9 of 10 to 7.4 of 10 (p < 0.001). Overtreatment recommendations decreased from 48% to 16% (p < 0.001), and undertreatment did not change (28% to 23%; p = 0.20) with the addition of photos.

CONCLUSIONS

The addition of wound photos to existing data as available via chart review and telephone consultation with patients significantly improved diagnostic accuracy and confidence, and prevented proposed overtreatment in scenarios without SSI. Post-discharge mobile health technologies have the potential to facilitate patient-centered care, decrease costs, and improve clinical outcomes.

Patients with Risk Factors for Complications Do Not Require Longer Antimicrobial Therapy for Complicated Intra-Abdominal Infection

A prospective, multicenter, randomized controlled trial found that four days of antibiotics for source-controlled complicated intra-abdominal infection resulted in similar outcomes when compared with a longer duration. We hypothesized that patients with specific risk factors for complications also had similar outcomes. Short-course patients with obesity, diabetes, or Acute Physiology and Chronic Health Evaluation II ≥15 from the STOP-IT trial were compared with longer duration patients. Outcomes included incidence of and days to infectious complications, mortality, and length of stay. Obese and diabetic patients had similar incidences of and days to surgical site infection, recurrent intra-abdominal infection, extra-abdominal infection, and Clostridium difficile infection. Short- and long-course patients had similar incidences of complications among patients with Acute Physiology and Chronic Health Evaluation II ≥15. However, there were fewer days to the diagnosis of surgical site infection (9.5 ± 3.4 vs 21.6 ± 6.2, P = 0.010) and extra-abdominal infection (12.4 ± 6.9 vs 21.8 ± 6.1, P = 0.029) in the short-course group. Mortality and length of stay was similar for all groups. A short course of antibiotics in complicated intra-abdominal infection with source control seems to have similar outcomes to a longer course in patients with diabetes, obesity, or increased severity of illness.

Patients with Risk Factors for Complications Do Not Require Longer Antimicrobial Therapy for Complicated Intra-Abdominal Infection

A prospective, multicenter, randomized controlled trial found that four days of antibiotics for source-controlled complicated intra-abdominal infection resulted in similar outcomes when compared with a longer duration. We hypothesized that patients with specific risk factors for complications also had similar outcomes. Short-course patients with obesity, diabetes, or Acute Physiology and Chronic Health Evaluation II ≥15 from the STOP-IT trial were compared with longer duration patients. Outcomes included incidence of and days to infectious complications, mortality, and length of stay. Obese and diabetic patients had similar incidences of and days to surgical site infection, recurrent intra-abdominal infection, extra-abdominal infection, and Clostridium difficile infection. Short- and long-course patients had similar incidences of complications among patients with Acute Physiology and Chronic Health Evaluation II ≥15. However, there were fewer days to the diagnosis of surgical site infection (9.5 ± 3.4 vs 21.6 ± 6.2, P = 0.010) and extra-abdominal infection (12.4 ± 6.9 vs 21.8 ± 6.1, P = 0.029) in the short-course group. Mortality and length of stay was similar for all groups. A short course of antibiotics in complicated intraabdominal infection with source control seems to have similar outcomes to a longer course in patients with diabetes, obesity, or increased severity of illness.

A Prognostic Model of Surgical Site Infection Using Daily Clinical Wound Assessment

BACKGROUND

Surgical site infection (SSI) remains a common, costly, and morbid health care-associated infection. Early detection can improve outcomes, yet previous risk models consider only baseline risk factors (BF) not incorporating a proximate and timely data source-the wound itself. We hypothesize that incorporation of daily wound assessment improves the accuracy of SSI identification compared with traditional BF alone.

STUDY DESIGN

A prospective cohort of 1,000 post open abdominal surgery patients at an academic teaching hospital were examined daily for serial features (SF), for example, wound characteristics and vital signs, in addition to standard BF, for example, wound class. Using supervised machine learning, we trained 3 Naïve Bayes classifiers (BF, SF, and BF+SF) using patient data from 1 to 5 days before diagnosis to classify SSI on the following day. For comparison, we also created a simplified SF model that used logistic regression. Control patients without SSI were matched on 5 similar consecutive postoperative days to avoid confounding by length of stay. Accuracy, sensitivity/specificity, and area under the receiver operating characteristic curve were calculated on a training and hold-out testing set.

RESULTS

Of 851 patients, 19.4% had inpatient SSIs. Univariate analysis showed differences in C-reactive protein, surgery duration, and contamination, but no differences in American Society of Anesthesiologists scores, diabetes, or emergency surgery. The BF, SF, and BF+SF classifiers had area under the receiver operating characteristic curves of 0.67, 0.76, and 0.76, respectively. The best-performing classifier (SF) had optimal sensitivity of 0.80, specificity of 0.64, positive predictive value of 0.35, and negative predictive value of 0.93. Features most associated with subsequent SSI diagnosis were granulation degree, exudate amount, nasogastric tube presence, and heart rate.

CONCLUSIONS

Serial features provided moderate positive predictive value and high negative predictive value for early identification of SSI. Addition of baseline risk factors did not improve identification. Features of evolving wound infection are discernable before the day of diagnosis, based primarily on visual inspection.

A patient-centered system in a provider-centered world: challenges of incorporating post-discharge wound data into practice

OBJECTIVE

The proposed Meaningful Use Stage 3 recommendations require healthcare providers to accept patient-generated health data (PGHD) by 2017. Yet, we know little about the tensions that arise in supporting the needs of both patients and providers in this context. We sought to examine these tensions when designing a novel, patient-centered technology - mobile Post-Operative Wound Evaluator (mPOWEr) - that uses PGHD for post-discharge surgical wound monitoring.

MATERIALS AND METHODS

As part of the iterative design process of mPOWEr, we conducted semistructured interviews and think-aloud sessions using mockups with surgical patients and providers. We asked participants how mPOWEr could enhance the current post-discharge process for surgical patients, then used grounded theory to develop themes related to conflicts and agreements between patients and providers.

RESULTS

We identified four areas of agreement: providing contextual metadata, accessible and actionable data presentation, building on existing sociotechnical systems, and process transparency. We identified six areas of conflict, with patients preferring: more flexibility in data input, frequent data transfer, text-based communication, patient input in provider response prioritization, timely and reliable provider responses, and definitive diagnoses.

DISCUSSION

We present design implications and potential solutions to the identified conflicts for each theme, illustrated using our work on mPOWEr. Our experience highlights the importance of bringing a variety of stakeholders, including patients, into the design process for PGHD applications.

CONCLUSION

We have identified critical barriers to integrating PGHD into clinical care and describe design implications to help address these barriers. Our work informs future efforts to ensure the smooth integration of essential PGHD into clinical practice.