Patient Safety Focus

By Sharon M. Geaghan, MD

The Risks and Benefits of Home Use Tests

Several trends are driving home use testing, including an aging population and wider adoption of care models utilizing patient-­centered medical homes. The latter are aimed at providing continuous and coordinated medical care, led by a physician, physician's assistant, or nurse practitioner. There is also increasing consumer demand for home-based devices, which offer convenience and lower cost compared to hospital or clinic services.

A Safety Risk?
Safety is as imperative in home health settings as in professional healthcare settings. An important point is that not all home testing requires a doctor's order. Home tests potentially risk bypassing physician care, inappropriate testing, self-diagnosis, self-treatment­, testing isolated from clinical context, and misinterpreted results.

At the same time, many studies demonstrate that patients can do quite well. The Food and Drug Administration (FDA) approves home use tests based on evidence that participants can properly use the test kits, generate accurate results, and correctly interpret the results. FDA offers guidance on optimizing home use tests (1), and has approved home use tests for glucose, cholesterol, fecal occult blood, HIV, menopause, ovulation, pregnancy, prothrombin time, and vaginal pH. Drugs of abuse and alcohol testing are also available.

A success story in home use testing is monitoring of anticoagulation therapy by point-of-care (POC) prothrombin time and international normalized ratio (PT/INR). These devices are not only safe and efficacious, but also have been shown to improve patient outcomes.

In the early 2000s, multisite U.S. and Canadian hospital-based anticoagulation centers reported successful pilot trials of home use POC PT/INR testing. Home testing offered a convenient alternative to the requirement for frequent trips to a hospital or clinic setting, often only for one test.

In 2008, Medicare released a national coverage policy supporting home use of FDA-approved portable INR monitors for patients with atrial fibrillation. Requirements included a physician's prescription, warfarin anticoagulation therapy for at least 90 days, and face-to-face training from a healthcare professional (2).

Since then, self-testing has become an even more compelling alternative to conventional testing. A single-center randomized French study compared conventional management of patients receiving oral anticoagulation by vitamin K antagonists with a self-testing program to determine reliability of the devices, and variability of self-measured INRs (3). Patients with mechanical heart valve placement either had INR monitoring at a laboratory at least monthly, or performed self-testing INR weekly, plus once-monthly clinic testing using one of two testing platforms for a mean of 49 weeks. The researchers found that the proportion of time spent within the target range was significantly higher in self-testers compared to the clinic testing group. In fact, adverse events were significantly higher in the clinic testing group compared with the patient self-testing group.

Unfortunately, despite good evidence of effectiveness, self-testing and self-management of oral anticoagulation has not been widely adopted. Investigators addressed this practice gap in a large meta-analysis that ­combined 11 trials, with data for more than 6,000 participants and 12,800 person-years of follow-up (4). A significant reduction in thromboembolic events in the self-monitoring group was evident, especially for patients younger than 55 years and participants with mechanical heart valves. Rates of major hemorrhage and death were not changed. Even the very elderly showed no significant adverse effects. Self-monitoring and self-management of oral coagulation are safe options for suitable patients of all ages, and patients should be offered this option to self-manage their disease with healthcare supervision.

Other Concerns
In addition to the evidence supporting the clinical utility of home use testing for certain conditions, there can be financial benefits for patients. For individuals without insurance, or those with high-deductible insurance, home tests are more affordable, even with an out-of-pocket expense (5). Self-monitoring blood glucose for type 2 diabetes was the first well-established home-based testing program and is reimbursed by Medicare and private payers. More than two-thirds of diabetics practice regular testing at home (6).

For both patients and the healthcare system to achieve the most ­benefit from home use testing, it is critical that home testing data be integrated with the electronic medical record. The personal health record (PHR) is one solution for capturing home test results. The PHR is a record that an individual or a caregiver maintains on a secure website. Providers, health plans, and private companies offer PHRs, and some are available for free (see myPHR.com). The PHR does not have to be tied to a particular health system and is controlled by the patient, rather than the institution. Widespread adoption of PHRs would increase integration of important test results from home-based testing with institutions' electronic medical records. We have a lot of work to do to get there.

The Future of Home Testing
Home monitoring will supplant some conventional intermittent laboratory testing practices in the near future. Automated implant technologies allow for continuous monitoring at home with remote result review. For example, vendors are racing to develop and launch real-time glucose monitoring technology.

Home testing can improve access to care, improve patient outcomes, and support programs in underserved areas. Stewardship of home use testing requires consideration of clinical and economic outcomes, and cost effectiveness of the whole patient episode within the healthcare delivery system.

References

1. The Food and Drug Administration. How you can get the best results with home use tests. http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/InVitroDiagnostics/HomeUseTests/ucm125664.htm. (Accessed August 1, 2014).
2. The Centers for Medicare and Medicaid Services. Decision memo for prothrombin time (INR) monitor for home anticoagulation management (CAG-00087R). http://www.cms.gov (Accessed September 2014).
3. Azarnoush K, Camilleri L, Aublet-Cuvelier B, et al. Results of the first randomized French study evaluating self-testing of the International Normalized Ratio. J Heart Valve Dis 2011;20(5):518–25.
4. Heneghan C, Ward A, Perera R, et al. Self-­monitoring of anticoagulation: A systematic review and meta-analysis of individual patient data. Lancet 2012;379:322–34.
5. Matthews AW. Worried about cholesterol? Order your own tests. WSJ Jan 11, 2011.
6. Centers for Disease Control and Prevention. ­Diabetes public health resource. http://www.cdc.gov/diabetes/statistics/incidence/fig3.htm (Accessed January 24, 2014).

By Michael Astion, MD, PhD
Editor-in-Chief, Patient Safety Focus

Presenteeism Among Healthcare Workers: A Negative Effect on Patient Safety

Over the years, Patient Safety Focus in Clinical Laboratory News has covered issues that are neglected in laboratory trade journals and scientific publications. These topics have included disruptive behavior, problems in specimen transportation, and disclosure of harmful errors to patients. This month I'd like to deal with presenteeism, which refers to coming to work sick.

Presenteeism received its name to contrast it with absenteeism, which refers to problems caused by not coming to work. The literature on presenteeism is dwarfed by the literature on absenteeism. In addition, discussions on loss of worker productivity dominate presenteeism literature, rather than the equally important effect of presenteeism on patient safety.

The facts about presenteeism for healthcare workers are troubling. Healthcare workers, including physicians and nurses, are more likely to come to work sick than other workers. They report a variety of illnesses while at work, including infectious diseases—most frequently respiratory or gastrointestinal infections—as well as depression and musculoskeletal complaints. There are a number of published case studies of patient harm caused by healthcare workers coming to work and spreading infectious diseases such as norovirus, ­methicillin-resistant Staphylococcus aureus, and influenza.

Why do healthcare workers come to work sick and why more than other workers? The answer is complicated, but a list of common themes has emerged. First is the feeling of being irreplaceable. Healthcare workers report feeling that patients would be harmed if they did not come to work and do their jobs. Thus, coming to work sick represents for them a patient safety tradeoff. Work culture is another important factor in presenteeism. Many healthcare workers report feeling they would be letting coworkers down if they stayed home sick, or that their work culture is not supportive of staying home. Other factors mentioned by healthcare workers that favor presenteeism are financial incentives to retain their store of paid time off, and feelings, especially among those with depression, that they need the structure of work to be stable.

The solutions for reducing presenteeism are poorly studied but there is a reasonable consensus around some pragmatic measures. These include unrestricted paid sick leave of at least 3 days, meaning no need for certification from a medical doctor. Other worthwhile interventions include screening employees during infectious diseases outbreaks and enforcing back-to-work criteria, such as specifying the number of symptom-free days before resuming work.

A number of healthcare care facilities have also started dealing with presenteeism by making it easier and less expensive for workers to obtain healthcare, including mental health services, at or near their facility.

Taken as a whole, these interventions are practical ways to support the health of healthcare workers and take on a sinister but neglected patient safety problem. Changing the work culture that underlies presenteeism requires strong leadership over a long period of time.

---

An interview with Christine Schmotzer, MD

Defining Critical Value Lists and Limits: How can labs balance efficiency and patient safety?

Defining, identifying, and rapidly communicating critical values is essential to the quality of care. But as the workload in clinical laboratories continues to increase and physicians face information overload, laboratories are forced to be more efficient without compromising patient safety. In this interview, Christine Schmotzer, MD, discusses how to design a critical value list and steps that labs can take to balance efficiency with patient safety. Schmotzer is the medical director of clinical chemistry at University Hospitals Case Medical Center and an assistant professor of pathology at Case Western Reserve University School of Medicine in Cleveland, Ohio.

Jaime Noguez, PhD, of the Patient Safety Focus editorial board conducted this interview.

Q What is the best strategy for establishing a critical value list and limits?
A Despite the importance of critical values in patient care and the emphasis on effective communication of these results in the past decade, there is no widely accepted guideline for defining which analytes should be on a critical value list and how the ­cutoffs should be assigned. Developing a critical value list remains at the discretion of each institution. In practice, a common group of tests—including glucose, potassium, hemoglobin, hematocrit, and platelets—appear on the critical value list of nearly every institution. The specific values for these commonly covered analytes, as well as other analytes that should be included beyond the common ones, vary considerably between institutions. The best strategy for your lab is to use all available data to guide your decision. This includes published literature, peer comparisons, local institutional data—especially the populations being served—and the local opinion and consensus of clinicians working at your institution.

Ideally, labs would use outcomes literature to determine cutoffs at which a specific analyte value becomes life-threatening if an intervention is not taken. But outcomes literature is limited, due in part to the challenges of obtaining broadly applicable data in varied patient populations. A number of surveys and institutional case studies have been published on this topic emphasizing the institutional variability of critical value lists/cutoffs and the lack of a well-defined mechanism for establishing them (1–4). The availability of these surveys and studies allows laboratories to compare their lists to others and provides insight into whether your institution is over or under-restrictive in critical value calling. Survey data should be transferred with caution as it may not be current, and may not be a suitable match to your patient population.

Another approach to improving your critical value list is specific peer-to-peer comparisons. Peer comparison can enable a lab to select institutions with a similar patient mix and complexity of population which may lead to critical value lists that are more directly transferrable or comparable to your own institution. Peer comparison has been enabled in the last decade by widespread availability of current institutional critical value lists and cutoffs on the Internet (5–7). These are provided by national labs, university-based labs, and other hospital labs. In general, it is relatively easy to find a peer, either through the Internet or your professional network.

Regardless of your initial approach to data-gathering, developing a critical value list and cutoffs should include discussion among clinicians, nurses, laboratory directors, and staff representing various departments and specialties. It is in this setting that institution-specific practices and needs can be discussed and influence the critical value list. For example, if an institution performs all blood gases at the point-of-care rather than in a decentralized laboratory, it may not need pH or pO2 to be included on a critical value list. Without specific outcomes literature to guide decisions, institutional and personal experience can be solid guides to setting critical value cutoffs that best meet the needs and philosophy of an institution.

Q Are there any other factors that need to be considered when designing your critical value list?
A While literature review, peer comparisons, and consulting with your physicians are important, assessing your current state, including critical result distribution, call frequency, and reporting logistics can provide insight into opportunities for improving your critical call list and process. Determining the tests leading to the highest number of calls and the units receiving the most calls can lead to valuable insights. For example, we were surprised to find critical vancomycin levels were in our top 10 most called tests. Further exploration led to practice changes to enhance the relationship between time of draw and drug administration, as well as discussion on whether abnormal vancomycin levels met the definition of a critical—immediately life-threatening—value. An important but often overlooked factor in successful critical value policies and procedures is the capabilities of your laboratory information system (LIS) for helping you identify and flag critical results. Many LISs don't have the ability to assign unit-specific flags. For example, clinical consensus at your institution may show that the threshold for critically low potassium can be different for inpatients versus outpatients. If your LIS does not allow for different critical results based on inpatient or outpatient status, the critical result will likely be set at the most conservative cutoff.

Q How can labs improve their critical values notification efficiency without compromising patient safety?
A The art of critical value policy and procedure is to call all life-threatening results while not inundating clinicians with information they already know and slowing lab flow with unnecessary calls. Keeping the focus on the definition of a critical result as a test result which is immediately life-threatening can contribute to a list that maximizes both efficiency and patient safety. One example of this is reviewing the critical value list, looking for test results which can strongly impact patient care if overlooked, but do not need immediate or urgent intervention. An institution may consider re-classifying them as semi-urgent or notification values and create a separate list with its own requirements for notification time frames. An example would be a positive HIV test result or abnormal cystic fibrosis sweat test. The risk to a patient if this result is missed is significant, but it is not immediately life-threatening. Again, the LIS and institutional processes may or may not be able to handle separate tiers of result notification.

Other strategies to balance efficiency and patient safety involve working with clinical leadership to eliminate calls to units in which the critical result is expected, such as abnormal troponins to cardiac ICU or elevated creatinine and BUN to the dialysis unit. Similarly, clinical leaders will often support eliminating repeat calls for select analytes which were previously critical within 24 hours, such as white blood cell and platelet counts.

As with any process geared toward improving patient safety, communication is critical. The more interaction and discussion a laboratory has with its clinicians on establishing a critical value list and clinically-appropriate reporting processes, the greater the impact on overall patient safety.

References

1. Wagar EA, Friedberg RC, Souers R, et al. Critical values comparison: A College of American Pathologists Q-Probes survey of 163 clinical ­laboratories. Arch Pathol Lab Med 2007;131:1769–75.
2. Howanitz PJ, Steindel SJ, Heard NV. Laboratory critical values policies and procedures: A college of American Pathologists Q-Probes Study in 623 ­institutions. Arch Pathol Lab Med 2002;126:663–9.
3. Dighe AS, Rao A, Coakley AB, et al. Analysis of ­laboratory critical value reporting at a large academic medical center. Am J Clin Pathol 2006;125:758–64.
4. Lum G. Critical limits (alert values) for physician notification: Universal or medical center specific limits? Ann Clin Lab Sci 1998;28:261–71.
5. Mayo Medical Laboratories: Critical values/critical results list. http://www.mayomedicallaboratories.
   com/articles/criticalvalues/view.php?name=Critical+Values%2FCritical+Results+List (Accessed August 2014).
ARUP Laboratories: Critical values lists. http://www.aruplab.com/Testing-Information/resources/PDF_Brochures/ARUP_Critical_Values.pdf (Accessed August 2014).
6. University of Washington, Department of Laboratory Medicine: Critical values list. http://depts.washington.edu/labweb/PatientCare/Clinical/Critical.htm (Accessed August 2014)