Evaluating the Connections Between Primary Care Practice and Clinical Laboratory Testing: A Review of the Literature and Call for Laboratory Involvement in the Solutions

Our primary goal is to identify potential laboratory testing performance and quality indicators in the preanalytic and postanalytic phases of testing. As an initial step, through funding from the Centers for Disease Control and Prevention, we performed a literature search to assist in the development of performance and quality indicators that might serve to improve primary care laboratory testing processes important for quality and safety.

A focused search of peer-reviewed literature, without meta-analysis, was developed in order to establish the current state of knowledge regarding laboratory-related processes involving clinical laboratories and primary care offices and the resulting errors. The literature scan was developed through consultation with a librarian at the University of Colorado in Aurora specializing in database and Internet information science, and used Ovid/MEDLINE as a primary source of information. Supplemental information was gathered from several conference proceedings and multiple electronic bibliographic databases, such as PubMed. The search strategy used for the bibliographic databases was a combination of title and topic that included the following primary domains and alternative terms: patient harm (medical error, adverse events, and patient safety), laboratory tests (results, diagnostic tests, and screening tests), communication (transition, handoff, handover), and primary health care (ambulatory care, family practice, internal medicine, general practice, and patient-centered medical home). The Agency for Healthcare Research and Quality Advances in Patient Safety publications from prior Agency for Healthcare Research and Quality–funded studies were manually searched for related articles, as they are not included in the above databases. The search included the following inclusion/exclusion criteria: (1) limited to literature published from the last 10 years, (2) available in English, and (3) no global restrictions applied.

Articles selected for review underwent further analysis for the level of evidence quality by 2 reviewers, using a grading scale developed by the Centers for Disease Control and Prevention called Developing Systematic Evidence Review and Evaluation Methods for Quality Improvement.23 Briefly, the scale required the reviewers to rate the quality of each article based on a 10-point scale, while evaluating the following 6 dimensions: observed effect size, relevance, study, practice, outcome measures, and findings/results. Articles that received greater than 7 points on average were reviewed by the 8 reviewers as a group. Major themes from each article were identified and associated with the affected IOM quality domains.

All articles were first reviewed by title and abstract, yielding a total of 95 potentially relevant articles. Ten articles scored greater than 7 points on the grading scale and were reviewed in their entirety by the group of reviewers. Each article was discussed by the group and key points were extracted. A summary of the key points from all the articles revealed the following 7 key concepts in laboratory medicine–primary care: (1) process failures, (2) delays, (3) communication gaps, (4) errors in judgment and cognition, (5) influence of minorities/language, (6) practice culture, and (7) lack of patient centeredness. Table 1 identifies and defines the main themes from the articles along with the associated IOM quality domains and affected TTP phases. Table 2 displays the proposed potential metrics based on the themes found in the literature review.

Highlighting an important overall theme, several articles addressed breakdowns or failures in the overall TTP. First, some practices demonstrated a complete lack of a process for ordering, implementing, and reporting laboratory results. Elder et al24 observed 2 primary care practices that had no written protocols or procedures for any results management steps. Similarly, Pace et al25 also determined that a total lack of a process system was associated with laboratory errors and a frequent contributor to patient harm. Lin et al26 reported that difficulty accessing laboratory results and the lack of a defined follow-up system contributed significantly to lost laboratory results. Hickner et al27 reported that practices that used more than one laboratory, and thus were forced to have more than one process, were much more likely to experience preanalytic and postanalytic errors in testing (17% versus 5%). In addition, Hickner et al27 found failures in all phases of the TTP. Casalino et al28 reported that partial medical records were highlighted as potential causes of error because having information in multiple areas added complexity to the process. The end result of process failure is clinical decision without necessary laboratory information. Smith et al29 identified 6.1% of primary care visits were performed with missing laboratory results. Graham et al30 studied mitigation of near-miss events and found a high number of physician-mitigated events in the preanalytic phase of the TTP after failures in the laboratory process. Lack of or failures in a process may affect all IOM quality domains, but are most significant in the domains of effectiveness, timeliness, and efficiency.

Four articles specifically addressed how delays in reporting were a contributing factor to laboratory errors. Staes et al31 found that delays in reporting and processing significantly increase risk to patients due to delays in decision making by a patient's clinical team. In addition, when laboratory tests are delayed or missed, patients incur additional costs, inconvenience, and discomfort for repeat testing without additional benefit. Twenty-four percent of errors resulted in delays in diagnosis. Hickner et al,27 Pace et al,25 and Gandhi et al12 all reported that postanalytic delays in reporting to physicians and to patients can contribute significantly to laboratory related errors. Delayed diagnoses primarily affect the IOM quality domain of safety and timeliness.

Preanalytic and postanalytic communication gaps were addressed in 3 articles and were identified as a major theme in the literature. Gandhi et al12 reported that communication handoffs contributed to 20% of laboratory errors and represented a primary factor that lead to errors. Similarly, Pace et al25 identified that communication within the practice and between the practice and another location can be subject to communication gaps and errors that lead to patient harm. Lin et al26 described how communication gaps may result in communication failures among staff and found that 16% of staff stated in a survey that they were not clear on who was responsible for following up on laboratory results. Communication failures focus on the IOM quality domains of safety, communication, and patient centeredness.

Two articles addressed the impact of preanalytic and postanalytic errors in judgment or cognition. Gandhi et al12 reported that frequent breakdown points occurred when providers failed to order appropriate diagnostic tests (55%) and when physicians misinterpreted a diagnostic test (37%). The most common explanation for these failures was that the physician seemed to lack knowledge of the appropriate test. Other factors, including failures in judgment (79%) and vigilance or memory (59%), also contributed to errors. Likewise, Pace et al25 reported that judgment was a significant factor that contributed to laboratory errors that lead to patient harm. The IOM quality domains of safety and effectiveness are highlighted by judgment and cognitive failures.

Two articles commented specifically on how race/ethnicity influences laboratory errors. Hickner et al27 reported that errors in laboratory test implementation were nearly double for minority groups compared with non-Hispanic whites (32% versus 18%). Additionally, minority patients were more than twice as likely to experience adverse consequences or harm as a result of the errors. Similarly, Pace et al25 associated language barriers with harm to patients with respect to laboratory medicine. These studies highlight the IOM quality domain of equity.

The culture of a practice regarding laboratory testing and laboratory errors was a major theme identified by the reviewers and was evaluated in 2 articles. Kaprielian et al32 described how a common barrier to reporting and thus addressing laboratory errors is fear of betraying colleagues. Elder et al24 stated that safety awareness was an important factor in assessing test results management quality and that leadership focus and communication that occurs around quality and safety was necessary. A safe culture impacts all phases of testing and is important for IOM quality domains of safety, efficiency, and patient centeredness.

Hickner et al27 commented on establishing explicit rules for the postanalytic notification of results and eliminating the use of the phrase “no news is good news”—an assumption that the Agency for Healthcare Research and Quality counsels patients not to make.33 Graham et al30 echoed this focus on patient centeredness, commenting that offices should empower patients to contact them if things do not appear to be handled correctly. They continued to add that patients have the potential to be effective mitigators, as they have a different lens on events and can see things that are often missed by office staff.

The comprehensive literature review documented the following themes as sources of laboratory–primary care errors: breakdowns in process, delays in diagnosis, communication gaps, errors in judgment and cognition, influence of minorities and speech barriers, practice culture, and patient centeredness. Many articles suggested each event was the result of not one but a series of errors. These themes are fertile fields for the development of practice metrics aimed at evaluating primary care–laboratory medicine risk and decreasing risk of error and adverse events. A summary of findings, IOM domains involved, and potential metrics is given in Table 1.

In an ideal state, laboratories and primary care practices have a duty to their patients to develop comprehensive processes designed to allow accurate and complete follow-up beginning in the preanalytic phase and ending in the postanalytic phase. Two articles (Staes et al31 and Lin et al26) recommended the use of computerized reporting systems in order to improve process tracking. However, others suggest complex technology-driven systems may result in overreliance (Hickner et al27) and simple paper tracking systems may be superior (Kaprielian et al32). As a start, a potential metric may be an outlined procedure for laboratory medicine process in practice protocols.

The lean principle of just-in-time work flow suggests customers receive what they want and need, exactly when they need it.34 As customers of our services, patients deserve timely and accurate results. Laboratories should work with practices to develop more efficient ways to communicate results in an effort to ease and improve the communication of results to patients. As a proposed benchmark, laboratory results should be communicated to patients within one business day of receipt by the practice.

Two areas of communication failure were identified: failures within primary care practices and failures between practices and laboratories. One important lean manufacturing principle is to standardize work, which includes identifying who is responsible for each step of a task.34 Practices might assign one support staff member to track all laboratory orders and results or spread the responsibility, as long as it is clear who is responsible for what. Again, laboratories have an opportunity to improve their service by working closely with practices to improve communication.

This might seem like one of the more difficult themes to address and will require more study. However, root cause analysis35 of the fundamental errors in laboratory utilization may reveal additional factors contributing to errors. Perceived errors in judgment and cognition often result in individual blame, finger pointing, and retraining when in reality these errors may be due to underlying process design failures. Attempts should be made to separate true failures in judgment and cognition from system-related failures. One of these systematic failures is the limited exposure to laboratory medicine in medical school and residency. With the ever-burgeoning volume of medical science to keep up with, perhaps primary care providers need more access to skilled laboratorians to help advise them on appropriate testing algorithms and result interpretation. Electronic physician order entry systems have the capability of including reminders and remarks to assist with judgment and cognition when placing laboratory orders. It is unlikely that simply asking primary care physicians to work harder on laboratory result tracking, as has been suggested,28 will improve outcomes, particularly without associated incentives. As a metric for study, we postulate that primary care providers should have convenient access to laboratory physicians to assist in laboratory test ordering and interpretation. It should be noted that appropriate reimbursement for these efforts is a limiting factor in realizing this additional support.

Our literature highlighted an important concept with regard to minority and language barrier populations and their increased risk in the laboratory–primary care testing process. This important area requires additional study as to the root cause of this finding and possible solutions. One may speculate cultural and language barriers lead to isolation and difficulties communicating regarding tests ordered, results, and significance.

The importance of practice culture with regard to safety cannot be overemphasized. Without a global mission for patient safety and quality and support from leaders in the practice and laboratory, significant improvements are not likely. Metrics to measure practice culture are very difficult to develop. Indirect measures include the presence of a quality improvement program, monthly quality conferences, procedures on how to handle errors, and whether error tracking is performed.

As our ultimate customers, patients play an important role in the process. Studies reviewed showed that patients are effective mitigators of near-miss events and should be more actively involved in the process. Practices should have a protocol for informing patients of laboratory results and involving them in the decision-making process.

Primary care providers are customers of the clinical laboratories, because they generate samples for testing. Therefore, directors of clinical laboratories play a role in addressing their customer needs and providing timely and accurate service. The literature review highlights several areas wherein the clinical laboratory could strive to improve its service, not just in the analytic phases of testing but extending service to the preanalytic and postanalytic phases. Based on the findings from the literature review, several possible metrics have been proposed to begin the evaluation of the primary care laboratory process (Table 2). Our next steps involve practice survey studies and finalization of possible performance metrics.