The Emergence of Lab-Pharmacy Partnerships

Much has been written about how laboratory professionals can partner with clinicians to improve patient outcomes. But as physicians struggle to keep up with the demands of today's complex healthcare system and the constant stream of new research findings, some forward-thinking laboratorians are finding that working directly with pharmacists can enhance the value and efficiency of both lab tests and drugs.

This trend has gathered momentum, especially over the last few years, according to Paul Jannetto, PhD, DABCC, FACB, MT (ASCP). "I've noticed that labs in many institutions are becoming partners with the pharmacy, so that now it's not just the lab and the physician, but a triad with the lab, physician, and the pharmacist working together to make sure patients' medication levels are being monitored and adjusted appropriately," Jannetto said. "I think this is part of a cultural shift where the lab is becoming less isolated and more integrated into the hospital and the healthcare system." Jannetto is an associate professor of pathology at the Medical College of Wisconsin, and director of the Clinical Chemistry/Toxicology Lab, the Rapid Response Lab, and co-director of the heme-oncology lab at Dynacare Laboratories/Froedtert Hospital in Milwaukee.

First the Script, then the Lab

When it comes to the lab's role in pharmacotherapy, most of the attention lately has been paid to pharmacogenomics, where genetic tests help predict whether a drug is suitable for a patient, or at what dose therapy should begin. But for some time, experts in patient safety have advocated another kind of personalized medicine—that is, close monitoring of patients on new medications, as well as continued monitoring as long as the patient takes the drug.

For certain drugs like anticoagulants, intensive monitoring with regular lab testing is routine and necessary to avoid adverse drug events (ADEs). But patients on many other types of drugs also could benefit from close monitoring, according to Gordon Schiff, MD, associate director at the Center for Patient Safety Research and Practice at Brigham and Women's Hospital in Boston.

"People talk a lot about appropriate drug selection, but I think it's equally important that patients be monitored carefully," Schiff said. "This takes on a number of dimensions, and one of the most important is monitoring lab results." Schiff also is associate professor of medicine at Harvard Medical School, and co-investigator at both the University of Illinois College of Pharmacy in Chicago and the Brigham and Women's Hospital Centers for Research in Therapeutics, which have worked on lab-pharmacy linkage projects.

In a recent review article focused on older adults, Schiff and his co-authors argued that many, if not most, ADEs result from a lack of close monitoring rather than improper initial prescriptions (J Am Geriatr Soc 2011;59:1513–1520). Schiff and his co-authors pointed out that in the U.S., 54% of ADE-related hospitalizations in older adults can be attributed to drugs that require regular monitoring. The authors also noted that 16% of ADEs resulted from clinical lab results being ignored, and that monitoring-related ADEs were almost 50% more frequent than ADEs resulting from inappropriate initial prescriptions. Perhaps not surprisingly, physicians have trouble with lab results used to monitor drugs, with almost 60% expressing dissatisfaction with this aspect of their practice.

Schiff and his co-authors emphasized that while the data on laboratory monitoring is very limited, the studies that have explored this area suggest that problems are common, especially for chronic conditions. For example, 37–55% of patients taking digoxin did not have serum digoxin concentrations checked at least once per year, 12–63% of patients taking angiotensin-converting enzyme inhibitors had inadequate monitoring of potassium and creatinine, and 37–61% of patients taking valproic acid or carbamazepine did not receive an annual complete blood count.

So, what's the solution? According to Schiff and his co-authors, the problem calls for a "framework of team-based approaches to patient management" based on lab and pharmacy collaboration. "With this approach, prescribing quality does not hinge on static attributes of the initial prescribing decision, but entails a dynamic process in which the benefits and harms of drugs are actively monitored, managed, and reassessed over time," the authors wrote.

Managing High-Risk Drugs

In the same way that lab menus have grown, the list of drugs that require therapeutic drug monitoring (TDM) has expanded too, requiring more laboratorians to become experts in this area. Moreover, some of the newer drugs require a more sophisticated approach, according to Bill Clarke, PhD, MBA, DABCC, associate professor and director of therapeutic drug monitoring and toxicology in the department of pathology at Johns Hopkins University School of Medicine in Baltimore. In this environment, it makes sense for laboratorians and pharmacists to have closer ties.

Some drugs, such as busulfan, used in cancer treatment, or mycophenolate, used in organ transplantation, require more than just a single measurement with a simple cutpoint, Clarke noted. "For these types of drugs, you really need a multidisciplinary team beyond the lone physician," he said. "Physicians depend on input from the lab and pharmacy because the breadth of medical knowledge required to care for these patients is just growing exponentially. What clinicians really need are decision support tools. That's where the lab and pharmacy can work together to help."

At Johns Hopkins, for example, physicians lean heavily on specialized oncology pharmacists for busulfan monitoring. The pharmacist, in turn, coordinates closely with the lab. Used to destroy bone marrow and cancer cells in preparation for a bone marrow transplant, the drug has a narrow therapeutic range, with potential toxic side effects such as life-threatening liver damage. Samples from up to 10 different time points may be taken for a single infusion of the drug, and the pharmacist must analyze the pharmakinetics for the infusion based on an area under the curve (AUC) analysis.

"This is a labor intensive process that requires a lot of interaction among me, the technologists running the tests, and the pharmacist who is making the dose adjustment," Clarke said. "Of course, the clinicians know this is going on in the background, but they do not need to be involved every step of the way."

Another drug that requires monitoring, voriconazole, is an anti-fungal used either to treat aspergillosis or as prophylaxis in an immunocompromised patient before bone marrow transplantation. The lab at Johns Hopkins measures levels of the drug three times a week, and Clarke often has follow-up discussions with specialized infectious diseases or oncology pharmacists when levels are higher than expected.

Such collaboration springs in part from a trend toward distributed ownership for clinical care that results in broadened roles for laboratorians as well as pharmacists, Clarke explained. "Pharmacists have become more specialized, and in hospitals, they're taking an active role in the pharmacologic analysis of patients," he said. "In addition, physicians have less of an authoritarian view where they must attend to every detail. Given the nature of healthcare and how much there is to do, some of these responsibilities are getting distributed, and I think as part of that changing model, today there is more of a collaborative consulting role for the pharmacist and the laboratorian."

Clarke's view was echoed by Jannetto, whose teamwork with the pharmacy began when he represented the lab on the Froedtert Hospital interdisciplinary Pharmacy, Nutrition, and Therapeutics (PNT) committee. The committee manages the institution's formulary, but also monitors appropriate utilization, safety, and efficacy of medications. For example, Froedtert recently switched to measuring free phenytoin versus total in hospitalized patients to get a more accurate picture of toxicity and therapeutic effects. Before the PNT committee recommended this change, Jannetto made sure the lab had the assay validated on multiple platforms so that when the switch occurred, testing would not be interrupted and would be available around the clock. Jannetto worked with his pharmacist colleagues outside the PNT committee as well, adjusting reference intervals for TDM assays when necessary based on evidence-based guidelines. Jannetto most recently updated Froedtert's reference intervals for the antibiotic vancomycin based on literature that showed that the therapeutic range should be raised to prevent resistant organisms. In addition, Jannetto consults with pharmacists about which TDM assays need to be implemented internally versus being sent-out to reference labs.

The key is forging a two-way relationship. "The pharmacists may need certain TDM results in a timely fashion to facilitate dosing adjustments for patients or the expansion of the lab's test menu for new medications being added to the hospital formulary," Jannetto said. "At the same time, they have a better understanding than most physicians of all the preanalytical variables and can help us make sure samples are drawn at the right time and the results are interpreted appropriately."

Tackling Chronic Disease with Teamwork

While the lab-pharmacy collaboration for TDM has gained increased visibility, progress on monitoring patients with chronic diseases has been slower. Once patients are out of the hospital, it's often up to individual physicians to follow patients on their own. Integrated delivery systems, however, have an advantage in this area.

One example that offers a glimpse into the interconnected health systems many expect to grow in the future is underway at Group Health Cooperative (GHC) in Seattle, Wash. This healthcare system is taking chronic disease monitoring to a new level with a unique lab-pharmacy partnership. Beyond intensive TDM, the idea is to systematically crosscheck lab and pharmacy data to make sure patients do not develop side effects or other problems with long-term prescriptions. For example, GHC recommends that patients taking adalimumab for rheumatoid arthritis have complete blood count, creatinine, and alanine transaminase tests every 2 months.

Laboratorians, pharmacists, and physician representatives in the GHC system collectively review the literature and agree on the list of drug-lab pairings and the frequency of lab testing. As part of the process of refilling prescriptions for these mediations, GHC pharmacists check that patients have had applicable lab testing done at prescribed intervals (See Table, below).

Pairing Drugs and Tests for Better Monitoring The following are examples of the more than 50 drugs that make up the monitoring program at Group Health Cooperative in Seattle, Wash.
Medication	Tests	Frequency
Amiodarone (Cordarone)	TSH, AST or ALT	6 Months
Azathioprine (Imuran)	CBC, AST or ALT	3 Months
Divalproex (Depakote), Valproic Acid	Valproic acid level, CBC, AST or ALT	6 Months
Lithium (Lithobid)	Lithium level, serum creatinine,TSH	6 Months
Atypical antipsychotics Clozapine Risperidone Olanzapine Quetiapine Ziprasidone Aripiprazole	HbA1C or fasting glucose, Fasting lipid profile	Annual
NSAIDS	Creatinine for patients over age 70	Annual
Beta Interferons (Avonex, Betaseron)	CBC, ALT or AST, Bilirubin	6 Months
Methotrexate	CBC, AST or ALT, Creatinine	3 Months
Oxcarbazepine (Trileptal)	Sodium	6 Months
Riluzole (Ritutek)	ALT or AST	3 Months
Tizanidine (Zanaflex)	ALT or AST	6 Months
Topiramate (Topamax)	Bicarbonate, Creatinine	Annual
Alpha 1-antitrypsin (Prolastin)	LFT, BUN, Creatinine	Annual
Adalimumab (Humira)	CBC, Creatinine, AST or ALT	2 Months
Anakinra (Kineret)	CBC, Creatinine, AST or ALT	3 Months
Etanercept (Enbrel)	CBC, Creatinine, AST or ALT	2 Months
Sulfasalazine (Azulfidine)	CBC, PLT, Creatinine, AST or ALT	3 Months
Mercaptopurine (Purinethiol)	CBC, LFT	3 Months
5ASA (Asacol, Pentasa, Dipentum)	LFT, Creatinine	6 Months
Cyclosporine (Sandimmune, Neoral, Gengraf)	CBC/Diff, Chem 14, Magnesium, Cyclosporine A	3 Months
Tacrolimus (Prograf)	CBC/Diff, Chem 14, Magnesium, Tacrolimus	3 Months
Sirolimus (Rapamune)	CBC/Diff, Chem 14, Magnesium, Sirolimus	3 Months
Lamivudine 100 mg Adefovir 10mg Entecavir 0.5 mg Entecavir 1.0 mg	CBC, LFT, BUN, Creatinine	3 Months
Zafirlukast (Accolate)	AST or ALT, Bilirubin	6 Months
Marrow stimulants; erythropoietin, darbepoietin	CBC or Hemoglobin or Hematocrit	3 Months
Oral corticosteroid (If prednisone tablet average daily dose is less than 10 mg the patient is excluded)	HbA1c or fasting glucose	12 Months
Abbreviations: ALT, alanine aminotransferase; AST, aspartate transaminase; BUN, blood urea nitrogen; CBC, complete blood count; Chem 14, comprehensive metabolic panel; PLT, platelet count; LFT, liver function test panel; TSH, thyroid-stimulating hormone.

The program originated in 2003 as a way to improve efficiency for both the pharmacy and the lab. Since its inception, the number of drugs requiring monitoring with lab tests has grown significantly, and more growth is expected in the future, said Jim Carlson, PharmD, GHS director of pharmacy health plan services. "As more and more medicines require monitoring, not only for the initial prescription, but ongoing monitoring of the patient's care, this program will continue to grow," he said. "Now, as new drugs come onto the market, we have a process in place."

Carlson's lab partner on the project, GHC's chief of clinical laboratories, Kim Riddell, MD, emphasized the value of working together with pharmacy to offer more comprehensive care. Before this program was implemented, there was little standardization of which drugs needed monitoring with which lab tests, she explained. Physicians would use their best judgment, but with considerable variation. In addition, an unknown number of patients were not being followed-up. "With this program, we're monitoring patients in a manner that is consistent throughout the organization," Riddell said. "It's really about implementing best practices across the board."

While not all labs are in a position to launch a program quite like that at GHC, which operates as both a provider and insurer, Riddell expressed confidence that many labs could find success with similar initiatives, though on a smaller scale. "I do think there is a wider role for laboratory and pharmacy cooperation," Riddell said. "Most labs know their physicians, which patients they serve, and where the patients come from. They also probably have a good idea of the major pharmacy players in their area. So I think at least an educational effort with doctors, patients, and pharmacists would be worth while." Riddell suggested labs start small, with commonly prescribed drugs that generate the most testing, and explore whether there is common ground with local pharmacists and providers.

To engage pharmacists, Carlson advises labs to delineate clearly the goal of improving the clinical outcomes for patients. "This is what will really speak to the pharmacy folks," he said. "They went into pharmacy in order to help patients, and my feeling is that most would be open to hearing from other healthcare professionals about collaborative opportunities to actually improve patient care."

Next: Data Integration

Critical to GHC's success in monitoring medications is the organization's having given pharmacists electronic access to lab data. As more organizations use electronic medical records (EMRs), such access will likely become more widespread. In a review article, Schiff has argued that providers should push these connections even further—making sure both the lab and pharmacy see a patient's entire record and creating alerts for physicians based on lab-drug pairs (Arch Intern Med 2003;163:893–900).

Echoing the model Carlson and Riddell have adopted at GHC, Schiff emphasized that physicians need systems that actively assist them in medication monitoring. "It's not enough to say that we have recommendations that you should, for example, monitor liver function tests if a patient is prescribed a certain drug," he said. "We need to go further—how often exactly should they be tested, how do you respond to results, how do we notify physicians and patients when they're failing to get these tests? I think there is quite an important future for this."

Schiff also noted that labs can benefit from greater connectivity to patients' mediation lists to aid in interpreting tests. For example, with delta checks, the lab could use a medication list to hone in on what truly is an unexpected result in the case of thrombocytopenia: the condition could be medication induced and a new finding, or the patient may already be under treatment. "Even if the physician has not kept up the problem list, the medication list can quite quickly let the lab know what that result means. This helps the lab interpret the result in a way that is more informed, and it helps labs and physicians act in a way that's more efficient at both ends."

In the future, Schiff would like to see greater adoption of computerized physician order entry alerts that arise automatically from predetermined lab-drug pairs. In this area, a lot of ground remains to be covered, he contended. "We've seen some small steps in terms of systems that are somewhat interoperable and where there can be decision support rules written that link to both lab and pharmacy data. But by and large, there is still a long way to go."