Process for developing this document

Leadership from the Society for Healthcare Epidemiology of America identified the need for the task force and recruited members to lead the task force (D.D. and D.M.). They then publicly called for volunteers for the Diagnostic Stewardship Task Force with the purpose of developing a series of papers better define diagnostic stewardship in healthcare settings. Experts were chosen to represent varied expertise from diverse settings. During an initial meeting, experts discussed their views on current gaps in diagnostic stewardship, and priority areas. Subsequently, subgroups were formed based on interest and area of expertise.

The diagnostic pathway

The primary objectives of diagnostic stewardship are to improve patient care by promoting accurate and timely diagnosis and thereby increasing appropriate antimicrobial use while reducing antimicrobial resistance (Fig. 1).^{Reference Dik, Poelman, Friedrich, Niesters, Rossen and Sinha1,Reference Morgan, Malani and Diekema2,Reference Watson, Trautner and Russo17} Additional secondary benefits of diagnostic stewardship may include improving efficiency of care, reducing costs, and improving institutional metrics related to hospital-acquired infections.^{Reference Madden, Weinstein and Sifri6,Reference Bhowmick, Kirn and Hetherington18,Reference Pliakos, Andreatos, Shehadeh, Ziakas and Mylonakis19}

The diagnostic pathway is a process that starts and ends with clinicians who order a diagnostic test and make patient-care decisions based on their interpretation of the test result. Along the way, key steps include collection of a specimen for testing, processing the specimen, performing the test, and providing a result, all within the context of the institutional systems in place (Fig. 2). Opportunities for optimizing test use exist in each of these steps and must be carefully considered to ensure the success of a diagnostic stewardship interventions and avoid unintended consequences. We describe each step in more detail and provide a summary of diagnostic stewardship strategies and specific interventions in Tables 2 and 3, respectively.

Fig. 2. Conceptual model of diagnostic stewardship and interventions at different steps in the diagnostic pathway.

Table 2. Strategies and Concepts Used in Diagnostic Stewardship

Table 3. Targets for Diagnostic Stewardship for Common Microbiologic Tests

Note. PCR, polymerase chain reaction.

^a Microbiology laboratories include recommendations on specimen collection and transport (eg, transport device, preservative or not, transport temperature, time from specimen collection to test) in their procedure manuals, which are available to medical personnel.

^b CDC recommends reporting of Candida auris in suspected cases. C. auris screening is recommended for patients with overnight stay in a hospital outside of teh United States in the previous 1 year.

Diagnostic stewardship strategies

Most diagnostic stewardship interventions focus on commonly used diagnostic tests that are major drivers of antibiotic use, such as urine or respiratory cultures.^{Reference Sullivan13} Other areas of attention include improving use of expensive tests that rarely influence clinical decision making (eg, respiratory virus panel for immunocompetent individuals or whole genome sequencing in the routine work-up of meningoencephalitis) and frequently obtained low-cost tests that contribute to unnecessary spending, particularly when false-positive results can lead to patient harm from additional diagnostic procedures or treatment (eg, avoiding blood cultures for isolated fever without concern for sepsis). Tests are sometimes targeted if they are used to diagnose conditions that are publicly reported, such as C. difficile tests or tests that provide indeterminate results (eg, QuantiFERON-TB Gold, β-D-glucan)^{Reference Fabre, Shoham, Page and Shah42,Reference Fabre, Markou and DeMallie43} or lack sensitivity and/or specificity (eg, the West Nile virus nucleic acid amplification test lacks clinical sensitivity).^{Reference Karaba, Blair, Martin, Saheed, Carroll and Borowitz44} At a minimum, hospitals should develop strategies for optimal practices of blood cultures, urine cultures, respiratory cultures, and C. difficile testing given the implications for patients and hospitals (Box 1).

Diagnostic stewardship activities may rely on clinical-decision support tools, education, and testing and reporting-based strategies.^{Reference Morgan, Malani and Diekema2} The effectiveness of different diagnostic stewardship strategies varies widely and depend on numerous factors.^{Reference Mora, Krug, Grigonis, Dawson, Jing and Hammerman12,Reference Xie, Woods-Hill and King45} When considering a diagnostic stewardship intervention, one should consider not only reported effectiveness but also the feasibility of implementation based on local resources, workflow, and work preferences, as well as the burden the intervention may pose to clinicians. Some interventions are less intrusive, such as education and soft stops, but these require human resources and information technology (IT) support. The benefits of a diagnostic stewardship strategy must be carefully weighed against the potential for unintended consequences such as shifting to empiric treatment without testing. A summary of general diagnostic stewardship strategies is presented in Table 2, and a summary of potential interventions for common infectious syndromes and microbiologic tests is described in Table 3.

The impact of diagnostic stewardship interventions should be measured both for effectiveness and safety. For example, the effectiveness of an intervention to reduce inappropriate treatment of asymptomatic bacteriuria through implementation of an algorithm to reduce unnecessary urine cultures may be assessed by measuring antibiotics prescribed for “urinary tract infections” (outcome measure), the number of readmissions due to UTI or escalation of care due to missed UTI (safety measure), and frontline providers’ satisfaction with the new workflow (process measure).

More in-depth discussion of diagnostic stewardship strategies to improve antimicrobial use and hospital-acquired infection metrics and measures to evaluate the impact of diagnostic stewardship interventions will be included in a future guidance paper.

Teams and support required for diagnostic stewardship

Diagnostic stewardship efforts have been driven by a mix of motives including improving patient outcomes and antibiotic use, decreasing healthcare-associated infections, and improving cost-effectiveness. It is important that any diagnostic stewardship intervention targeting an infection be evaluated by a multidisciplinary team given the ramifications an intervention may have in terms of resources needed to carry out the intervention and the potential impact on patient management and hospital metrics (Box 2).^{Reference Dik, Poelman, Friedrich, Niesters, Rossen and Sinha1,Reference Patel and Fang9} For example, rapid molecular tests, which have improved pathogen detection (faster time to organism identification with identification of antimicrobial resistance through gene detection), only improved patient management in recent studies when implemented with support from an antimicrobial stewardship team that provides guidance on result interpretation and management decisions.^{Reference Bhowmick, Kirn and Hetherington18,Reference Cosgrove, Li and Tamma25,Reference Timbrook, Morton, McConeghy, Caffrey, Mylonakis and LaPlante26} End users should also be involved during the early stages of conception of a diagnostic stewardship intervention to ensure the proposed intervention does not increase clinical burden.

It is critical that those leading interventions have experience with clinical diagnosis and treatment as well as an understanding of the laboratory process, local processes for ordering, regulatory requirements, and quality improvement. Expertise in implementation and behavioral science is also useful to best design interventions that will change practice. Many hospitals have a laboratory utilization committee that could describe existing resources and structure to help develop and sustain diagnostic stewardship.

Many of the strategies discussed make use of the electronic medical record. IT support is important not only to “build” the intervention in the electronic medical record but also to track the impact of the intervention. Furthermore, diagnostic stewardship has not traditionally been supported financially as an independent entity in a healthcare system but as part of infection control, antimicrobial stewardship, or clinical microbiology. The time and resources required to effectively implement diagnostic stewardship changes, especially to champion the project at the unit level as well as monitor for intended and unintended consequences of interventions, can be significant, which highlights the need for support from hospital leadership. Clear leadership support can also improve frontline provider acceptance of diagnostic stewardship interventions. Many cost-saving opportunities are related to effective diagnostic stewardship, including reductions in direct costs from tests not performed and antimicrobials not prescribed and indirect costs from improved hospital-acquired infection–related metrics and reduced length of stay,^{Reference Watson, Trautner and Russo17,Reference Pliakos, Andreatos, Shehadeh, Ziakas and Mylonakis19,Reference Fleming, Hess and Albert46} which are relevant for making a business case to support diagnostic stewardship efforts.

Diagnostic stewardship beyond microbiology

The principles and goals of microbiology diagnostic stewardship can be applied to diagnostic tests outside microbiology, including radiology or other nonlaboratory diagnostics that affect the management of infections. Tests or evaluations that are used to diagnose infectious disease, such as chest radiographs, echocardiograms, and ophthalmologic evaluations, should be considered by diagnostic stewardship programs.

A full discussion of diagnostic stewardship for tests outside microbiology is beyond the scope of this document. Such a report would require the same approach as with infectious disease tests, including clinical evaluation by those with content expertise to evaluate the different steps in the diagnostic pathway for the test and consider potential benefits and harms to be measured.

Future needs for diagnostic stewardship

Attention to the problem of test overuse in medicine is increasing; however, the impact of inappropriate testing on patient harm has not been well studied. National management guidelines of infectious diseases often lack a diagnostic stewardship perspective.^{Reference Fabre, Spivak and Keller47,Reference Fabre, Carroll and Cosgrove48} Specific recommendations to improve use of common microbiologic tests, such as blood and urine cultures, have been published across age groups; however, many opportunities exist to further standardize recommendations for specific disease processes, patient populations, and testing practices.^{Reference Fabre, Carroll and Cosgrove48–Reference Achten, Klingenberg and Benitz54} High-risk patient populations, such as immunocompromised patients and patients living with medical devices, may also benefit from consideration of diagnostic stewardship strategies. Benchmarking has been implemented to improve patient safety in hospitals and significant efforts are ongoing to prevent healthcare-associated infections. Benchmarking of common microbiologic tests utilization does not exist but may have value as a diagnostic stewardship strategy (eg, establishing an inappropriate urine culture utilization rate for specific patient populations).^{Reference Fabre, Carroll and Cosgrove48,Reference Chen, Bilker, Hamilton, O’Donnell and Nachamkin55}

Data on optimal measures of effectiveness, safety, and implementation are limited.^{Reference Mora, Krug, Grigonis, Dawson, Jing and Hammerman12} In practice, quality improvement projects are frequently implemented with a focus on test reduction and lack an evaluation of patient outcomes.^{Reference Mora, Krug, Grigonis, Dawson, Jing and Hammerman12} Assessment of the full impact of diagnostic stewardship interventions is needed, including potential unintended consequences for both patients and clinicians.^{Reference Gupta, Boland and Aron56} Lastly, the success of intervention strategies may differ depending on the implementation approach. Future studies comparing approaches and assessing implementation outcomes may identify the most efficient and effective diagnostic stewardship strategies for adoption and sustainment of practice changes.^{Reference Proctor, Silmere and Raghavan57}

Electronic support tools have improved the use of diagnostic tests; however, there is still a great need to understand the effect of the healthcare work system and cultural and social drivers of inappropriate use of common diagnostic tests.^{Reference Xie, Woods-Hill and King45} Potential sociobehavioral strategies to address common mental models driving inappropriate test use have been described.^{Reference Vaughn, Szymczak, Newton and Fakih37}

Diagnostic stewardship will need to expand to address new and expensive or unapproved tests, such as multiplex molecular panels, 16S rRNA gene sequencing, metagenomic and whole genome sequencing, to tackle both existing and emerging infectious threats such as antimicrobial resistance. Institutional support for diagnostic stewardship efforts will be important for successful implementation and sustainability of diagnostic stewardship activities and should grow beyond individual hospitals to include different healthcare settings and systems.

In summary, diagnostic stewardship is an emerging method to improve patient care by better implementing common diagnostic tests. Most changes occur in the electronic health record or laboratory and can therefore be easily scaled to large populations of patients. Comprehensive approaches to diagnostic stewardship by expert teams can reduce diagnostic errors and lead to more accurate diagnoses and treatment, reduced healthcare costs, and decreased antibiotic resistance.