Article, Emergency Medicine

Kiosks as tools for health information sharing: exploratory analysis of a novel ED program

American Journal of Emergency Medicine 32 (2014) 797-810

Correspondence

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage: locate/ ajem

Kiosks as tools for health information sharing: exploratory analysis of a novel

ED program?,??

It is widely recognized that the demand for emergency services is rising. Emergency department (ED) utilization increased from 36.9 visits per 100 persons per year in 1995 to 42.8 in 2010 [1,2]. With increasingly time- and personnel-constrained EDs, self-service kiosks have emerged as one potential solution to perform simple tasks such as registration and information distribution. This process-oriented solution may free up health care providers for more complex duties and is particularly relevant to EDs, whose core mission is to provide emergency care for potentially life-threatening conditions.

A non-targeted opt-in rapid oral fluid HIV testing program has been part of standard of care in this ED since 2005 [3,4]. Recently, we evaluated the feasibility and outcomes of a kiosk approach to testing in order to improve efficiency and reach a greater proportion of patients [5]. As part of that quality insurance program evaluation, we included a structured questionnaire on the kiosk to determine patient comfort level with using kiosks to share and update personal health information with ED staff. This report presents a secondary analysis of data [5].

Kiosk-facilitated screening involved two stages, a front-end regis- tration kiosk to engage those interested in HIV testing, and a back-end testing kiosk that collected demographic data and provided a single location for testing [5]. The front-end kiosk surveyed patients about Comfort level with using kiosk technology, measured via a 5-point Likert scale (Figure). Level 1 Emergency Severity Index patients who were sent directly to treatment rooms and non-ambulatory patients were excluded from the program evaluation, due to inability to use the free-standing kiosks [6]. The study was approved by the Johns Hopkins University School of Medicine Institutional Review Board.

Between December 2011 and April 2012, 4351 patients completed the kiosk module. Table 1 summarizes patient demographic and clinical information collected from the electronic medical record. The majority of patients (57%) responded positively to using the kiosk, indicating that they felt either “very comfortable” (32%) or “some- what comfortable” (25%) with using it to update their information; 15% rated “neutral”, while 6% answered “not very comfortable” and 16%, “not at all comfortable” (Table 2). We performed a multivariate regression analysis to determine whether patient characteristics were associated with kiosk comfort level (Table 3). Men were less confident than women with using kiosks to enter information (OR, 0.8) and patients age 65 and older were less likely to express comfort utilizing kiosks for this purpose relative to those 18-24 years old (OR, 0.6).

To the best of our knowledge, our analysis is one of the first to examine patient preferences surrounding kiosk utilization for per- sonal data entry. Porter et al (2004) reported the use of a kiosk in ED Pediatric asthma cases. Parents entered information about their

? The authors do not have associations that might pose a potential conflict of interest.

?? Support was provided by Gilead Sciences, Inc.’s HIV FOCUS program. Hsieh,

Rothman, and Gaydos are also supported by NIH U54EB007958.

child’s illness experience and the kiosk provided parent-child needs and recommended actions to improve the child’s care. Parents’ responses to the asthma kiosk were extremely positive as most found it easy to navigate, characterizing the kiosk interaction as a valuable use of time, and appreciating the action item outputs [7].

Our study revealed age and gender discrepancies in level of comfort with using kiosks as men and elderly patients were less comfortable using kiosks to enter and update personal information. Although we did not specifically assess Logistical challenges associated with the module, we have reported previously that higher education levels and prior experience with kiosks are associated with less time spent on specific kiosk modules and higher patient-reported ratings of ease of use [8]. The 2008 US Census Bureau reports that 14.5% of men over age 18 in Baltimore City do not graduate from high school, as compared to 12.7% of women [9]. This may pose an intrinsic challenge to kiosk usage if men in certain locations on the whole are less educated and less comfortable with new technologies. In regards to elderly patients, it is likely that a certain amount of the discrepancy will disappear as self-service kiosks become more common in settings outside the hospital such as airports, grocery stores, banks, retail stores, etc. In order to be widely implemented as a tool for data entry and information disbursement, kiosks must be easy to understand and navigate for most patients.

Our findings suggest promise for use of kiosks in the ED as a supportive communication tool with the majority of patients expressing comfort with using the kiosk to share health information with providers. Self-service kiosks have the potential to speed up ED visits because patients can fill in medical and surgical history and Medication lists during wait times. This study provides an excellent foundation upon which to introduce kiosk interventions into an urban ED setting with the goals of optimizing ED throughput and improving information sharing.

Megan S. Orlando, BA

Department of Emergency Medicine The Johns Hopkins University School of Medicine

Baltimore, MD E-mail address: [email protected]

Richard E. Rothman, MD, PhD

Department of Emergency Medicine The Johns Hopkins University School of Medicine

Baltimore, MD Division of Infectious Diseases

The Johns Hopkins University School of Medicine

Baltimore, MD

Alonzo Woodfield, MD Megan Gauvey-Kern, MS Stephen Peterson, BS Peter M. Hill, MD, MSc

Department of Emergency Medicine The Johns Hopkins University School of Medicine

Baltimore, MD

0735-6757/(C) 2014

798 Correspondence / American Journal of Emergency Medicine 32 (2014) 797810

Figure. Screenshot with question about comfort using kiosks to update medical information.

Table 1

Baseline characteristics of all study patients (N = 4351)

Characteristic No. (%)

Age

18-24 1038 (23.9)

25-34 1102 (25.3)

35-44 698 (16.0)

45-54 880 (20.3)

55-64 450 (10.3)

N 64 182 (4.2)

Female sex? 2454 (56.4)

Race?

African American

3356 (77.1)

White

742 (17.0)

Other

253 (5.8)

Acuity?

1

2 (0.1)

2

17 (0.4)

3

3010 (69.2)

4

1124 (25.8)

5

67 (1.5)

Charlotte A. Gaydos, MS, MPH, DrPH

Department of Emergency Medicine The Johns Hopkins University School of Medicine

Baltimore, MD Division of Infectious Diseases

The Johns Hopkins University School of Medicine

Baltimore, MD

Yu-Hsiang Hsieh, MSc, PhD

Department of Emergency Medicine The Johns Hopkins University School of Medicine

Baltimore, MD

http://dx.doi.org/10.1016/j.ajem.2014.04.025

References

Chief complaint?

[1] Institute of Medicine. IOM Report: the future of emergency medicine in the United

Abdominal pain

466 (10.7)

States health system. Acad Emerg Med 2006;13:1081-5.

Headache

166 (3.8)

[2] National Hospital Ambulatory Medical Care Survey: 2010 Emergency Department

Back pain

175 (4.0)

Summary Tables; 2010.

Abscess

112 (2.6)

[3] Hsieh Y-H, Jung JJ, Shahan JB, et al. Outcomes and cost analysis of 3 operational

models for rapid HIV testing services in an academic inner-city emergency department. Ann Emerg Med 2011;58(1):S133-9.

Chest pain

112 (2.6)

Other

3320 (76.3)

Disposition?

Admit

380 (8.7)

Discharge

3333 (76.6)

  1. Hsieh Y-H, Jung JJ, Shahan JB, Moring-Parris D, Kelen GD, Rothman RE. Emergency medicine resident attitudes and perceptions of HIV testing before

Other (observation) 638 (14.7)

ED primary discharge diagnosis related to infectious disease? 398 (9.2)

Table 3

Regression results on patient comfort with using kiosks

* Gender, race, chief complaint, and disposition information was missing in 128

patients; triage acuity information was missing in 131 patients; ED primary discharge diagnosis information was missing in 147 patients.

Table 2

Patient comfort with reviewing and making corrections to medical and surgical history and medication lists on a kiosk (N = 4351)

Odds ratio (95% CI)

Age 18-24 1.0

Age 25-34 0.9 (0.8-1.1)

Age 35-44 1.2 (1.0-1.4)

Age 45-54 1.0 (0.8-1.1)

Age 55-64 1.1 (0.9-1.4)

Age N 64 0.6 (0.4-0.8)

Female sex 1.0

Male sex 0.8 (0.7-0.9)

Non-black NS

Black NS

Comfort level

Number (%)

CC: other

CC: abd pain

NS

NS

Very comfortable

1398 (32.1)

CC: headache

NS

Somewhat comfortable

1074 (24.7)

CC: back pain

NS

Neutral

635 (14.6)

CC: abscess

NS

Not very comfortable

249 (5.7)

CC: chest pain

NS

Not at all comfortable 674 (15.5)

NS, not significant; abd pain, abdominal pain.

Correspondence / American Journal of Emergency Medicine 32 (2014) 797810 799

and after a focused training program and testing implementation. Acad Emerg Med 2009;16(11):1165-73.

  1. Hsieh Y-H, Gauvey-Kern M, Peterson S, et al. Novel emergency department registration kiosk for HIV screening increases engagement of high risk patients. J Telemed Telecare 2014.
  2. Gilboy N, Tanabe P, Travers DA, Rosenau AM, Eitel DR. Emergency severity index, version 4: implementation handbook. AHRQ publication no. 05-0046-2. Rockville, MD: Agency for Healthcare Research and Quality; May 2005.
  3. Porter SC, Cai Z, Gribbons W, Goldmann D, Kohane IS. The asthma kiosk: a patient- centered technology for collaborative decision support in the emergency. J Am Med Inform Assoc 2004;11(6):458-68.
  4. Rothman R, Gauvey-Kern M, Woodfield A, et al. Streamlining HIV testing in the emergency department–leveraging kiosks to provide true universal screening: a usability study. Telemed E Health 2014;20(2):122-7.
  5. Survey 2008 American Community. Sex by age by educational attainment for the population 18 years and over; 2008.

Redefinition of diagnostic role of inferior vena cava ultrasonography in the identification of acute heart failure

To the Editor,

We have read with great interest the report of Yavasi et al [1] entitled “Monitoring the response to treatment of acute heart failure patients by ultrasonographic inferior vena cava collapsibility index,” in which the authors explore the role of sonographic measurements of Inferior vena caval diameters and inferior vena caval- collapsibility index (IVC-CI) (ie, measurement of Respiratory variation of IVC diameter) in the monitoring of Acute heart failure therapy in the emergency department.

Correct diagnosis and management of AHF is lifesaving in patients with dyspnea; however, differential diagnosis is often challenging especially among patients with severe respiratory distress and when respiratory illness coexists. Several studies have shown the utility of IVC-CI for establishing the differential diagnosis between cardiac and noncardiac causes of dyspnea [2-5], for evaluating the response to diuretic treatment [6,7] and for predicting the prognosis of AHF patients [8,9].

Inferior vena caval-collapsibility index is a rapid, bedside, and Noninvasive tool that gives valuable information in many settings, being expression of the volume status, of the right atrial pressure and of the central venous pressure [10,11]. cardiac failure is usually associated with fluid overload, so a high central venous pressure and therefore a low IVC-CI are expected. Indeed, the main studies show that a reduced IVC-CI below different cut-off values (15%, 20%, 33%, and 50%) is highly suggestive for AHF among acutely dyspneic patients. Kajimoto et al [5] even suggest an algorithm in which a high IVC-CI (above 50%) rules out the cardiogenic cause of dyspnea.

Nevertheless, these studies share an important limitation, such as the lack of the degree of patients’ dyspnea, which is, in our opinion, a crucial point. Indeed, the reduction of the IVC diameter is related not only to a reduction of the central venous pressure but also to the degree of inspiratory negativization in intrathoracic pressure [10,12]. During decompensated heart failure, the strong inspiratory efforts can markedly lower intrathoracic pressure, so we think that patient’s respiratory pattern can influence the IVC-CI. Starting from this hypothesis, we conducted a study to evaluate the diagnostic role of the IVC-CI in this setting correlating this parameter with the degree of patient’s dyspnea.

We retrospectively compared a group of 46 dyspneic patients who received a final diagnosis of AHF with a control group of 35 patients with non-AHF dyspnea. Ultrasonography assessment, conducted before the administration of therapy, included IVC diameter, IVC-CI, and Lung ultrasound. These measurements were compared between

the 2 groups and correlated with the degree of dyspnea; this last being defined as the following scale: (1) exerting dyspnea, (2) mild/ moderate dyspnea at rest (respiratory rate, b 25 per minute), and (3) severe dyspnea at rest using accessory respiratory muscles (respira- tory rate, >=25 per minute).

An IVC-CI of less than 50% had 57% sensitivity (95% confidence interval (IC), 43%-71%), 83% specificity (95% IC, 71%-95%), 81% positive predictive value (95% IC, 67%-95%), and 59% negative predictive value (95% IC, 45%-73%) in diagnosing AHF. However, AHF patients with degree 3 of dyspnea and rapid onset of symptoms (hours) had an IVC-CI at least 50% (P b .001 and .003, respectively), showing that IVC-CI ultrasound must be interpreted according to the clinical contest; in particular, patients with severe respiratory distress have an elevated IVC-CI, that is, an IVC-CI at least 50% does not exclude AHF diagnosis.

In their patients, Yavasi et al [1] found a mean collapsibility index of 22% in the enrolled patients before the administration of treatment, which is lower of that reported in previous report (19%) [6]. They hypothesize that this difference is probably due to the inclusion of patients with both right and left heart failure, but we think that also the severity of dyspnea could have a role in influencing the amount of patients with higher level of IVC-CI in spite of the condition of heart failure.

To conclude, we are aware of the usefulness of IVC-CI ultrasound, and this is true in many clinical settings above all the management of dyspneic patient, but we would like to highlight the need of an accurate interpretation of the information that this tool can produce, considering all the factors that can influence the changes of IVC diameter during the respiratory cycle, in particular, the respiratory pattern.

Alice Gianstefani, MD

Division of Internal Medicine Department of Medical and Surgical Sciences

University of Bologna

S. OrsolaMalpighi Hospital

Bologna, Italy E-mail address: [email protected]

Francesco Savelli, MD

emergency unit of Faenza Hospital

Faenza, Italy

Annagiulia Gramenzi, MD

Department of Clinical Medicine

University of Bologna

S. OrsolaMalpighi Hospital

Bologna, Italy

Enrico Zucconi, MD

Emergency Unit of Faenza Hospital

Faenza, Italy

Nicola Di Battista, MD

Area Critica Ospedale Privato Accreditato

Villa Nigrisoli Bologna, Italy

Raffaella Francesconi, MD

Emergency Unit of Faenza Hospital

Faenza, Italy

Mario Cavazza, MD Emergency Unit Department of Emergency

General Surgery and Transplants

S. OrsolaMalpighi Hospital

Bologna, Italy

http://dx.doi.org/10.1016/j.ajem.2014.04.016

Leave a Reply

Your email address will not be published. Required fields are marked *