10125652632794J Am Board Fam MedJournal of the American Board of Family Medicine1557-26251557-262522086819328611310.3122/jabfm.2011.06.100255NIHMS354827ArticleInformation Chaos in Primary Care: Implications for Physician Performance and Patient SafetyBeasleyJohn WMD12WetterneckTosha B.MD, MS3TemteJonMD, PhD1LapinJamie AMS2SmithPaulMD1Rivera-RodriguezA. JoyMS2KarshBen-TzionPhD*12Department of Family Medicine, UW-Madison School of Medicine and Public HealthDepartment of Industrial and Systems Engineering, UW-MadisonDepartment of Medicine, UW-Madison School of Medicine and Public Health Corresponding author: Ben-Tzion Karsh, PhD Industrial and Systems Engineering, UW-Madison, 1513 University Avenue, Room 3218, Madison, WI 53706, tel: 608-262-3002, fax: 608-262-8454, bkarsh@engr.wisc.edu15220121120112422012246745751Purpose

The purpose of this paper is to explore the concept of information chaos as it applies to the issues of patient safety and physician workload in primary care and to propose a research agenda.

Methods

We use a human factors engineering perspective to discuss the concept of information chaos in primary care and explore implications for its impact on physician performance and patient safety.

Results

Information chaos is comprised of various combinations of information overload, information underload, information scatter, information conflict, and erroneous information. We provide a framework for understanding information chaos, its impact on physician mental workload and situation awareness, its consequences, discuss possible solutions and suggest a research agenda which may lead to methods to reduce the problem.

Conclusions

Information chaos is experienced routinely by primary care physicians. This is not just inconvenient, annoying and frustrating; it has implications for physician performance and patient safety. Additional research is needed to define methods to measure and eventually reduce information chaos.

Primary Health CareInformation Management/InformaticsComplexity ScienceMedical ErrorsPractice ManagementNational Center for Research Resources : NCRRUL1 RR025011-03 || RR
INTRODUCTION

Primary care is comprised of first contact care, longitudinal care, comprehensive care, and coordinated care.1, 2 Without information that is comprehensive, accessible, timely and correct, primary care physicians (PCPs) cannot deliver high quality care. Information problems do exist, 3-7 with an extrapolation from one study suggesting that errors related to information handling account for 29% of family medicine errors. 8 Because information is central to the effectiveness of primary care, we propose a conceptual framework to explain the relationship between information hazards, i.e. information problems that may contribute to errors, and PCPs’ ability to provide high quality care.

This framework is especially timely given the emphasis on electronic health record (EHR) use in primary care. The patient centered medical home (PCMH) movement encourages EHR implementation to improve information handling. However, EHR systems may not meet this goal. 9-11 Furthermore, in 2010 the “Meaningful Use” regulations brought incentives and penalties to promote EHR use. 12 EHRs can transform the quality, comprehensiveness, timeliness, and accessibility of information in both good and bad ways. 13 Both the move toward PCMH and the need to adopt EHRs will have effects on information needs and the design of support systems in primary care. It is therefore important to understand how information relates to primary care.

INFORMATION CHAOS

The vignette in Figure 1 illustrates five specific information hazards that arise in primary care and how they can lead to information chaos.

Information overload

Information overload occurs when there are too many data, e.g. written, verbal and non-verbal, physician’s memory, for the clinician to organize, synthesize, draw conclusions from, or act on. Primary care involves the integration of an increasing number of patient problems while taking into account family and community factors,1, 14, 15 The PCP care coordination role further adds to the volume of information PCPs must process and act on. 2, 16, 17 EHRs may make the situation worse by encouraging electronic copying and pasting, adding irrelevant information through use of templates, and the mixing of data needed for billing and legal protection with that needed for clinical care. One physician has called the problem “high fiber medical records” (Russo P, personal communication).

Information underload

Information underload occurs when necessary information is lacking. The lack of sufficient and available information is common in primary care.18 Problems include unavailable or incomplete records or poor recall by physician or patient.8 Patients may decide to not disclose sensitive information. At times, physicians are unaware that information is even available. For example, a physician may prescribe a “new” medication, only to find out later from the patient that it was already tried and stopped because of a side effect. Although this information may be in the chart, the physician may have not known it was there. Information that is theoretically “in the chart,” but not readily accessible, is for all intents and purposes, invisible.

Information scatter

Information scatter refers to having information located in multiple places. In any given primary care encounter, the needed information may be located on the intake form, in the paper chart or EHR, at the clinic of a consultant or a hospital, or in the patient’s or the clinician’s mind. While EHRs may reduce the problem of information scatter,19 they can also worsen the problem due to inadequate search methods to find information quickly and effectively.13

Information conflict

Information conflict occurs when the clinician is unable to determine which data are correct. Often the clinician is confronted with a variety of conflicting data regarding an issue. For example, the patient may think he or she is on one medication, the clinic record suggests a different one, and the hospital discharge note states yet another.

Erroneous information

This is when the information is wrong. For example, the patient tells the doctor they are taking aspirin for pain when in fact they are taking acetaminophen or information is entered into the wrong patient chart.4. When incorrect information gets into a record it may be hard to purge it or it may simply get replicated by clinicians without re-validating it with the patient.20

We conceptualize these five information hazards: information overload, underload, scatter, conflict and erroneous information, as information chaos. Separately or together, these hazards increase the risk of an information-related error and are experienced by primary care clinicians on a daily basis. While the concept of information chaos resonates with PCPs, the nature of information chaos and the factors that can affect the relationship between it and patient outcomes have not been explored in primary care and are likely underappreciated by clinicians, administrators, EHR programmers and planners. Information chaos is more than inconvenient, annoying and frustrating; there are operational implications that can impair physician performance, increase workload and reduce the safety and quality of care delivered.

CONSEQUENCES OF INFORMATION CHAOS

The field of human factors engineering and ergonomics (HFE)21-24 contributes to our understanding of information chaos. HFE studies and applies information about human cognitive and physical limitations, and human abilities to the design of systems, including, for example, work processes, tools, technologies, and environments, and in turn how systems impact human cognitive and physical performance. From HFE science, we hypothesize a conceptual framework for the way information chaos relates to the cognitive processes in primary care. This is presented in Figure 2.

Information chaos imposes demands on PCPs. Because of information overload, they may not be able to adequately review all the information and may miss important details. Because of information underload, scatter, and conflict, PCPs must expend effort searching for needed information through charts, phone calls, further questions to the patient and perhaps unnecessary testing. Erroneous information may lead to re-work, errors in diagnosis, or treatment decisions that threaten patient safety. The work PCPs do in response to information chaos is not value-added to the visit; it is only necessary to deal with the chaos. Importantly, it distracts the PCP from the primary work of the visit: caring for the patient.

From a cognitive point of view, information chaos, may contribute to two unwanted outcomes: impaired situation awareness (SA)25-30 and increased mental workload.31-33 SA is defined as a person’s awareness and understanding of his/her task-related situation. It has three levels: (1) perception of elements in the environment, e.g., patient cues/stimuli (pulse, color), chart, EHR, nurse, (2) comprehension of the meaning of those elements by integrating the disparate pieces of information and determining what is important and (3) projection of future status so that decisions can be made.25, 34 Whether or not accurate SA is achieved is dependent on the timing, comprehensiveness, accessibility, and quality of the information obtained. That means information chaos can directly impair SA. If SA is poor, it can result in impaired decision making.25, 26, 29, 30, 35

A related variable is mental workload, which is the amount of cognitive resources required for a task. Mental workload is a function of demand, which may be increased due to information chaos, and available resources, such as time.31, 33 Mental workload and SA are related. Poor SA increases mental workload, and high mental workload can degrade SA.36 High physician mental workload occurs when the mental demands imposed on the physician exceeds the physician’s resources cope with those demands. Both the poor SA and the high mental workload generated by information chaos can impair cognitive work29, 31, 33, 35 (e.g., diagnosis and treatment decisions), and may have negative impacts on safety.35

At the start of an office visit a PCP will have some SA, but it is incomplete. The clinician only knows what he/she remembers from previous visits, from a brief look at the patient’s chart, and possibly from a short meeting with the nurse who roomed the patient. As soon as the clinician enters the room, SA is dynamically updated based on sensory inputs such as how the patient looks, feels and sounds and from higher level processes such as communication with the patient and searching in the medical record. Whether or not accurate SA is achieved is dependent on the timing and quality of the information obtained through sensation, perception, communication and record searching. Because of information chaos, the clinician may never fully develop SA, and, due to information underload, may not even know they are not fully or accurately aware of the situation as is suggested by work in related fields. 37

During times of high mental workload which may be increased by information chaos, people involuntarily focus on fewer cues, considering fewer options and solutions because of cognitive tunneling.32 This occurs when people zoom in on a narrow set of cues or options because mentally they cannot handle more. In such situations, people are at risk for decision errors38 because they miss things they should have noticed such as patient symptoms, or abnormal lab results. In other words, in situations of high mental workload, people operate with selective and reduced capacity.32 Among PCPs, higher mental workload has been positively associated with a higher perceived probability of medical error.39

FACTORS AFFECTING INFORMATION CHAOS AND ITS CONSEQUENCES

The magnitude of the effects of information chaos on the quality of care delivered by PCPs is affected by a variety of moderators. We highlight three that affect the relationship: interruptions, physician expertise, and time.

Interruptions

Interruptions occur when something or someone in the environment unexpectedly disrupts the physician’s physical or mental workflow.40-42 This may shift the physician’s attention from the task he/she was originally engaged in, the primary task, to the interrupting task.43 Once this shift in attention occurs, memory of the primary task begins to decay in order to make room for the processes required to deal with the interrupting task.44 Many interruptions are unnecessary such as EHR systems that impede the flow of the physician-patient conversation.45 Interruptions may increase mental workload, reduce the ability to cope with information chaos and thereby potentially impair performance.

Expertise

Expertise is another variable that affects the relationship between information chaos and its consequences.46 It is probable that a PCP with more expertise can handle more information chaos, maintain better SA, and have less mental workload than a PCP with less expertise. With experience comes more accurate expectations about what is going on, where information resides, and what ambiguous or confusing information implies. Because of that, the same level of information chaos may have less impact on a more experienced clinician.

Time

Finally, the available time to cope with information chaos is important for patient safety. Physicians working for organizations that allow them to schedule patient visits with sufficient time to manage their patients’ problems will suffer fewer effects of information chaos; they can take the time to deal with the situation. On the other hand, when information chaos occurs in a time constrained environment (e.g. during a 15 minute visit for a patient with several chronic conditions and medications to manage), the impact on physician performance is exacerbated. Time is perhaps the most important resource available to deal with information chaos.

WHAT CAN BE DONE TO CALM THE CHAOS?

Reducing information chaos requires mechanisms to ensure that the necessary information is available and presented in a useable manner at the right time. Two general strategies have been proposed to calm the chaos. These include improved visit preparation and EHRs.

Visit Preparation

PCPs need to know the reason(s) for a patient visit. Clinics may have patients complete some type of “patient agenda” form, either electronic or paper, to be completed prior to the visit. Another approach is the teamlet47 or huddle,48 in which the nurse who rooms the patient meets briefly with the PCP to talk about the patient’s stated or written agenda. Utilizing more staff support to assist in obtaining and organizing information during the visit has been proposed as well.49 This is a good start, but at best lets the PCP know the agenda mere minutes before he/she enters the room and there may not be time to obtain needed information from external sources or review information in depth.

Electronic Health Records

EHRs contribute to more timely and available information but caveats exist, many of which have been discussed.11,45 EHRs are generally designed to facilitate data entry to conduct and document the process of care. As more and more data are available in an EHR, there is an even greater need for improved search methods and display techniques to present the data needed at the time of the patient visit. Ideal EHR design would allow relevant, needed information to be pushed to the clinician based on the reasons for the visit.

A RESEARCH AGENDA TO EXPLORE INFORMATION CHAOS IN PRIMARY CARE

Our hypothesis is that appropriate practice redesign will reduce information chaos and provide the PCP with needed situational awareness during all phases of the visit, allowing him/her make better care decisions, and accomplish more (and perhaps with less fatigue) than would otherwise be the case. More research is needed to determine just what this practice redesign should be.

We propose that if progress is to be made on controlling information chaos and its effects on physician performance, the following questions, at minimum, need to be addressed:

Which aspects of information chaos are most common and which ones pose the greatest threats to patient safety?

How can we measure and quantify the elements of information chaos or resulting behaviors to determine the success of an intervention?

What interventions can reduce the information hazards and lead to the greatest improvements in SA, mental workload and ultimately patient safety?

What are valid ways to measure situation awareness in the healthcare settings? Our team had to develop new measures since none existed specific to primary care and measuring it during actual patient care delivery is challenging.

What is the nature of the effects of the moderators? Can clinician expertise or patient scheduling help to address information chaos?

When is the right time and what is the right information to provide to PCPs so that they can prepare appropriately for a visit and reduce information chaos?

What is the role of the clinic staff, patients and caregivers to prepare information for a PCP visit? When is the right time to collect the information and how should it be presented?

The questions above are examples of what needs to be done to help improve primary care. Technology cannot alone solve the problems of information chaos, nor can the transformation to a PCMH. New thinking is necessary. We cannot accept that information chaos is “just the way primary care is.” If we do this, we are not taking advantage of primary care redesign and the power of information systems to maximize the quality and efficiency of patient care delivery.

Support: This study was supported in part by grants 1P20HS017115 from the Agency for Healthcare Research and Quality (PI: Karsh), 1R18SH017899 from the Agency for Healthcare Research and Quality (PI: Karsh), and 1UL1RR025011 from the Clinical & Translational Science Award (CTSA) program of the National Center for Research Resources National Institutes of Health.

Prior Presentations: Short talk at Wonca International Meeting, Cancun Mexico, May 20, 2010. 2 poster presentations.

Conflict of Interest: None

The published version of this article can be accessed on the Journal of the American Board of Family Medicine website at: http://jabfm.org/content/24/6/745.full

Human Subjects: Not applicable as no actual research on human subjects was conducted.

DonaldsonMYordyKLohrKPrimary Care America’s Health in a New EraWashington, D.C.National Academy Press1996StarfieldBPrimary CareJ Ambulatory Care Manage1993164273710128415ElderNCGrahamDBrandtEThe Testing Process in Family Medicine: Problems, Solutions and Barriers as Seen by Physicians and Their Staff: A Study of the American Academy of Family Physicians’ Research NetworkJournal of Patient Safety2006212532KuzelAJWoolfSHGilchristVJPatient reports of preventable problems and harms in primary health careAnn Fam MedJul-Aug20042433334015335132PhillipsRLDoveySMGrahamDElderNCHicknerJLearning From Different Lenses: Reports of Medical Errors in Primary Care by Clinicians, Staff, and Patients: A Project of the American Academy of Family Physicians National Research NetworkJournal of Patient Safety200621140146KuoGMPhillipsRLGrahamDHicknerJMMedication errors reported by US family physicians and their office staffQual Saf Health CareAug200817428629018678727MakehamMAStromerSBridges-WebbCPatient safety events reported in general practice: a taxonomyQual Saf Health CareFeb2008171535718245220DoveySMMeyersDSPhillipsRLA preliminary taxonomy of medical errors in family practiceQual Saf Health Care200211323323812486987NuttingPAMillerWLCrabtreeBFJaenCRStewartEEStangeKCInitial lessons from the first national demonstration project on practice transformation to a patient-centered medical homeAnn Fam Med20097325426019433844FernandopulleRPatelNHow the electronic health record did not measure up to the demands of our medical home practiceHealth Aff (Millwood)Apr201029462262820368591KarshBTWeingerMBAbbottPAWearsRLHealth information technology: fallacies and sober realitiesJ Am Med Inform Assoc201017661762320962121US. 42 CFR Parts 412, 413, 422Medicare and Medicaid Programs; Electronic Health Record Incentive Program; Final RuleServices DoHaH75Federal Register20104431444518SchiffGDBatesDWCan electronic clinical documentation help prevent diagnostic errors?N Engl J Med2010362121066106920335582BeasleyJWHankeyTHEricksonRHow many problems do family physicians manage at each encounter? a WReN studyAnn Fam Med2004240541015506571AbboEDZhangQZelderMHuangESThe Increasing Number of Clinical Items Addressed During the Time of Adult Primary Care VisitsJ Gen Intern Med2008231210581065BodenheimerTCoordinating Care -- A Perilous Journey through the Health Care SystemN Engl J Med2008358101064107118322289PhamHHO’MalleyASBachPBSaiontz-MartinezCSchragDPrimary care physicians’ links to other physicians through Medicare patients: the scope of care coordinationAnn Intern MedFeb172009150423624219221375SmithPCRodrigAGBublitzCMissing clinical information during primary care visitsJAMA200529356557115687311Institute of MedicineCrossing the Quality Chasm: A New Health System for the 21st CenturyWashington DCNational Academy Press2001ThielkeSHammondKHelbigSCopying and pasting of examinations within the electronic health recordInt J Med Informatics200776SS122128BeasleyJWHamilton-EscotoKKarshBHuman Factors in Primary CareCarayonPHandbook of Human Factors and Ergonomics in Patient SafetyMahwah, NJLawrence Erlbaum Associates2006921936SalvendyGHandbook of Human Factors and Ergonomics2John Wiley and Sons1997SandersMSMcCormickEJHuman Factors in Engineering and Design7New YorkMcGraw-Hill Science1993WickensCDLeeJDLiuYGordon-BeckerSAn Introduction to Human Factors Engineering2Upper Saddle River, NJPearson Education2004EndsleyMRTowards a theory of situation awareness in dynamic systemsHuman Factors1995373264HazlehurstBMcMullenCKGormanPNDistributed cognition in the heart room: How situation awareness arises from coordinated communications during cardiac surgeryJournal of Biomedical InformaticsOct200740553955117368112HoganMPPaceDEHapgoodJBooneDCUse of human patient simulation and the Situation Awareness Global Assessment Technique in practical trauma skills assessmentJournal of Trauma-Injury Infection and Critical CareNov200661510471052SalmonPMStantonNAWalkerGHJenkinsDBaberCMcMasterRRepresenting situation awareness in collaborative systems: A case study in the energy distribution domainErgonomics200851336738418311612SmithKHancockPASituation Awareness Is Adaptive, Externally Directed ConsciousnessHuman Factors1995371137148StantonNAStewartRHarrisDDistributed situation awareness in dynamic systems: theoretical development and application of an ergonomics methodologyErgonomicsOct20064912-131288131117008257HancockPAHoffmanJENational Research CouncilStress and cognitive workloadTactical display for soldiers: human factors considerationsWashington DCNational Academy Press1997129162HancockPASzalmaJLOperator stress & display designErgonomics in Design20031121318HancockPAWarmJSA Dynamic-Model of Stress and Sustained AttentionHuman FactorsOct19893155195372625347EndsleyMRGarlandDJSituation Awareness Analysis and MeasurementLawrence Erlbaum Associates2000StantonNAChambersPRGPiggottJSituational awareness and safetySafety Science2001393189204HancockPADesmondPAStress, Workload and FatigueCRC Press2000EndsleyMRJonesDGDisruptons, Interuruptions and Information Attack: Impact on Situation Awareness and Decision MakingPaper presented at: Human Factors and Ergonomics Society Annual Meeting Proceedings2001KahnemanDTverskyASlovicPJudgement under Uncertainty: Heuristics & BiasesCambridge, UKCambridge University Press1982TemteJGrasmickMO’HalloranPMental workload in primary carePaper presented at: AAFP National Research Network 8th Annual Convocation of Practices and NetworksMarch 7-9, 20082008Colorado Springs, COBrixeyJJRobinsonDJJohnsonCWJohnsonTRTurleyJPZhangJA concept analysis of the phenomenon interruptionANS Adv Nurs Sci2007301E264217299274ChisholmCDDornfeldAMNelsonDRCordellWHWork interrupted: a comparison of workplace interruptions in emergency departments and primary care officesAnn Emerg Med200138214615111468609PaxtonFHeaneyDHowieJPorterBAA study of interruption rates for practice nurses and GPsNurs Stand1996Jul10433336PotterPWolfLBoxermanSUnderstanding the cognitive work of nursing in the acute care environmentJ Nurs Adm2005357-832733516077274AltmannEMTraftonJGTask Interruption: Resumption Lag and the Role of CuesProceedings of the 26th annual conference of the Cognitive Science Society20044247ZhengKHaftelHMHirschlRBO’ReillyMHanauerDAQuantifying the impact of health IT implementations on clinical workflow: a new methodological perspectiveJ Am Med Inform AssocJul-Aug201017445446120595314HancockPAWeaverJLOn time distortion under stressTheoretical Issues in Ergonomic Science200662193211BodenheimerTLaingBYThe teamlet model of primary careAnn Fam Med20075545746117893389StewartEEJohnsonBCHuddles: improve your office efficiency in mere minutesFam Pract Manag2007272917598631SinskyCAImproving office practice: working smarter, not harderFam Pract ManagNov-Dec20061310283417139937AbbreviationsEHR

Electronic Health Record

HFE

Human Factors Engineering

PCMH

Patient Centered Medical Home

PCP

Primary Care Physician

SA

Situational Awareness

Clinical Vignette Illustrating Information Hazards

Information Chaos, Mental Workload, Situational Awareness, Moderators and Impacts