Feasibility and accuracy of medication checks via Internet video

Abstract

We investigated the feasibility and accuracy of using Internet-based videoconferencing for double-checking medications. Ten participants checked 30 different medications using a desktop PC and a webcam. The accuracy of the video-based checks was compared with ‘face-to-vial’ checks. The checks included the drug name, dosage and expiry dates of ampoules, vials and tablets, as well as graduations on syringes. There was 100% accuracy for drug name, dosage, and graduations on syringes greater than 1 unit. The expiry dates proved more difficult to read, and accuracy was only 63%. The mean overall accuracy was 91% for all items. Internet video-based medication double-checks may have a useful role to play in processes to ensure the safe use of medications in home care.

Introduction

Medication errors are the most common type of medical error and they occur frequently in the inpatient setting.^1,2 There is little documented evidence of the prevalence of medication errors in the community or home-care setting,³ but given the high incidence in the inpatient setting it is likely that mistakes do occur. The recognised method of reducing medication errors is independent checking by two qualified people, commonly referred to as conducting a ‘double-check’.^4–6 For the administration of potentially dangerous medications, such double-checking is standard practice in most inpatient facilities. In the community or home setting, it is usual for only one nurse to make a home visit, which eliminates the possibility of a double-check of medication immediately prior to administration.

We have previously investigated the potential of Internet-based video support for patients at home.^7–9 To our knowledge, using videoconferencing for double checking medication has not previously been explored. The aim of the present study was to assess the feasibility and accuracy of using Internet-based video communication for medication double-checking.

Methods

The study was conducted in two stages. In the first stage, the feasibility of reading a range of medication items using a web camera (webcam) was assessed. In the second stage, the accuracy of Internet-based medication checking was assessed by comparing the accuracy of video-based reading of medication items with readings conducted in person (‘face-to-vial’).

Feasibility

To accurately check medications, it is necessary to identify the small details on medication vials and the markings on syringes. Five webcams of different specifications were tested to determine whether they could capture legible images of medication items. The medication items comprised: (1) drug vials (with a label containing the drug name, concentration and expiry date. In some cases the expiry details were embossed on the vial, rather than printed) and (2) a dose drawn up in a syringe, with a graduated scale.

Freely available videoconferencing software (VID, Logitech) was used to link two computers over the Internet. Initial testing was conducted using a high definition auto-focus webcam (Pro9000, Logitech). A glass ampoule of heparinised saline (47 mm in length) was held in front of the webcam during the video call and an assistant attempted to identify the ampoule contents and the expiry date. We hypothesized that the auto-focus feature of the webcam would be beneficial for reading the small font (11 pt) of the ampoule label. The same experiment was repeated using four other webcams (listed in Table 1). The integrated webcams in the laptop computers provided the most legible image. The other cameras that were tested focused too far into the distance for the contents of the ampoule to be readable. The integrated webcams were selected for use in the accuracy study.

Table 1

Web camera specifications

Type of camera	Specification	Focus
Logitech C500	1.3 Mpixel	Fixed
Logitech Pro9000	2 Mpixel, Carl Zeiss optics	Autofocus
Toshiba integrated webcam	1.3 Mpixel	Fixed
Dell Vostro 3300 integrated webcam	2 Mpixel	Fixed
HP Mini 110 integrated webcam	0.3 Mpixel	Fixed

Accuracy

A convenience sample of ten volunteers participated in an accuracy study. The characteristics of the participants are summarised in Table 2. Participants were seated alone in a room with a desktop PC and a webcam. A research assistant was located in a separate room, with a laptop (with integrated webcam) and a range of medication items. These items comprised medications and syringes in various forms: 1 ml glass ampoules; 5 ml glass ampoules; 5 ml plastic vials; 10 ml plastic vials; a 100 ml intravenous solution bag; a 100 ml bottle of oral suspension; a sheet of tablets; a 1 unit syringe; a 2 ml syringe; a 5 ml syringe and a 10 ml syringe. Examples of medications items are shown in Table 3. A video call was made using Logitech VID between the participant and the research assistant. During the call, the research assistant sequentially held thirty medication items in front of the laptop webcam. For each item, the participants recorded the requested details of the medication on a standard data collection form. The participants could ask the research assistant to rotate the item, move it towards or away from the camera, but not to confirm the details of the item.

Table 2

Characteristics of volunteers

	Gender	Age (years)	Profession	Familiarity with medications	Familiarity with videoconferencing
1	Female	25–35	Business manager	No	Some
2	Male	18–25	Engineer	No	Yes
3	Female	35–45	Administration	No	Some
4	Male	45–55	Researcher	No	Yes
5	Female	25–35	Research assistant	No	Yes
6	Female	35–45	Nurse	Yes	Some
7	Male	35–45	Researcher	Yes	Yes
8	Male	45–55	Lecturer	No	Yes
9	Female	35–45	Nurse	Yes	No
10	Female	18–25	Student	No	No

Table 3

Type of observation and accuracy comparing video over Internet with face-to-vial check

Items checked	Example	No of observations	Video check – no correct (%)	Face-to-vial – no correct (%)
Printed drug dose /name (glass ampoule, bottle)	Figure 1	60	60 (100)	60 (100)
Printed expiry date (glass ampoule, bottle)	Figure 2	40	37 (93)	40 (100)
Plastic vial drug name	Figure 3	50	50 (100)	50 (100)
Plastic vial expiry date	Figure 4	40	25 (63)	40 (100)
Drug name/dose tablet silver backing	Figure 5	20	17 (85)	20 (100)
Drug expiry silver backing	Figure 6	20	17 (85)	20 (100)
Intravenous fluid solution name/concentration	Figure 7	20	20 (100)	20 (100)
Intravenous fluid bag expiry date	Figure 8	20	20 (100)	20 (100)
Syringe contents (unit syringe)	Figure 9	10	7 (70)	10 (100)
Syringe contents (ml syringe)	Figure 10	20	20 (100)	20 (100)
Total		300	273 (91)	300 (100)

On completion of the video-based medication check, participants repeated the process in person (‘face-to-vial’). To avoid any recall effects the in-person checks were conducted at a minimum of seven days after the video check.

Results

Video check

The ten participants each observed 30 items via video over the Internet, giving a total of 300 independent observations. The results of all items checked by each method are summarised in Table 3. The names of medications or solutions in vials, ampoules, bottles and intravenous bags, as well as the dosage or concentrations and amounts in syringes (ml) were accurately observed and recorded in 100% of cases by all participants. The amount in a unit syringe was correctly identified in 70% of cases, with 20% of the incorrect responses being within 1 unit of the correct value. The tablet name, dose and expiry date, printed on a silver backed packet were accurately identified in 85% of cases. Expiry dates, particularly for the embossed dates on plastic vials, proved more difficult to read (63% accuracy). The mean overall accuracy was 91% for all items.

Face to vial check

All ten participants recorded the 30 items with 100% accuracy in the face to vial check.

Discussion

The literature contains examples of the use of technology to assist with medication dispensing and pharmacy applications to ensure correct dosages are calculated. There are also reports of the use of videoconferencing with patients to ensure medication compliance.^10–13 To our knowledge, no other studies have examined the potential of using webcams and Internet-based real-time videoconferencing for medication double-checks.

Web cameras

By comparing different webcams, we were able to establish that using laptop computers with integrated webcams is feasible and accurate when checking medication items. Standard USB plug-in webcams are designed to focus on facial features, at a distance of about 30 cm from the camera, but not on an object close to the camera.¹⁴ At a closer distance, all of the USB webcams that we tested blurred the image, resulting in the drug name, dose and expiry date being illegible. Moving the ampoule away from and towards the camera sometimes allowed the ampoule to come into focus, although not consistently and not for long enough to be able to check the information.

All of the laptops with integrated webcams yielded legible images. These integrated webcams are designed to focus (and magnify) at a shorter range (i.e. on a face close to the laptop screen) than conventional external webcams. This feature assists in the examination of small medication items. Laptop computers with integrated cameras are also an attractive choice for potential home care situations because of their affordability, and ease of transport.

Challenges

As Internet performance varied by time and day, so image quality varied. During times of poorer quality video, the images were more difficult to read. In particular, embossed expiry dates on plastic vials could not always be reliably read. This limitation can however be addressed in a number of ways: (1) medications for home care are generally dispensed by a pharmacist and the checks in place at time of dispensing include expiry dates; (2) there is also the opportunity for checks of expiry dates to be performed prior to a clinician's departure to a patient's home and (3) sending still images via email may be useful.

Another challenge was establishing the optimum distance from the webcam to hold items. It was found that the ideal distance to hold an item from the laptop integrated camera was approximately 12–15 cm. It was beneficial to hold the item in front of a blank background to prevent the camera from focusing on the person holding the object. Difficult to read items, such as the embossed expiry dates on plastic ampoules, were made easier to read by tilting the lid of the laptop down, and thereby limiting direct overhead light. At times, the lighting in different rooms also affected the images and required adjusting.

Using these techniques a high accuracy in correctly identifying items using video over the Internet was achieved.

Rationale for double checks

In conducting this study a variety of items relevant to home care were tested. Many medications require reconstitution, or calculated portions of a vial, and these are commonly double checked in a hospital environment. A double-check procedure involves a second clinician to double-check: name, concentration of medication, expiry date, and amount drawn up in the syringe. All of these aspects of a double-check were tested for accuracy by videoconference. Except for the embossed expiry date and segments on a unit syringe, we established that participants could accurately check all aspects of a double-check in all cases. This could potentially enable a home-care clinician to prepare a medication in the home and be supported by a colleague through a video call over the Internet to double-check the medication preparation immediately before administration to a patient. Current practice for home-care medication administration does not allow a double-check. If double-checking is required, medication preparation is either outsourced to pharmaceutical companies which prepare individual syringes for each patient, or two clinicians are required to attend the home. Both of these options are time consuming and costly.

The present study had certain limitations. For example, it was conducted in a controlled environment, and may not be generalizable to other settings.

Conclusion

The prevalence of medication errors in the community and home setting is not well-documented and is likely to be underestimated. Where medication errors occur, they reflect a fault in the processes or systems that should be in place to prevent misadventure. Internet video-based medication double-checks may have a useful role to play in processes to ensure the safe use of medications in home care. In a controlled environment, we have demonstrated the feasibility and accuracy of webcam videoconferencing for medication double-checks. We have also identified the limitations of using this method to double-check medication preparations, particularly in relation to the checking of expiry dates. Further research is required to assess the effectiveness of this approach in the home environment.

Footnotes

Acknowledgements

We thank the Sporting Chance Cancer Foundation and the Royal Children's Hospital Foundation for support.

References

Kaushal

, Jaggi

, Walsh

, Fortescue

, Bates

. Pediatric medication errors: what do we know? What gaps remain. Ambul Pediatr 2004;4:73–81

Hughes

, Ortiz

. Medication errors: why they happen, and how they can be prevented. Am J Nurs 2005;105(Suppl. 3):14–24

Easton

, Morgan

, Williamson

. Medication safety in the community: a review of the literature. See http://www.nps.org.au/__data/assets/pdf_file/0008/71675/09060902_Meds_safety_June_2009.pdf (last checked 9 January 2012)

Weir

. Best-practice protocols: preventing adverse drug events. Nurs Manage 2005;36:24–30

Sullivan

, Buchino

. Medication errors in pediatrics – the octopus evading defeat. J Surg Oncol 2004;88:182–8

Levine

, Cohen

, Blanchard

, Guidelines for preventing medication errors in pediatrics. J Pediatr Pharmacol Ther 2001;6:427–43

Bensink

, Armfield

, Pinkerton

, Using videotelephony to support paediatric oncology-related palliative care in the home: from abandoned RCT to acceptability study. Palliat Med 2009;23:228–37

Bensink

, Armfield

, Russell

, Irving

, Wootton

. Paediatric palliative home care with Internet-based video-phones: lessons learnt. J Telemed Telecare 2004;10(Suppl. 1):10–13

Bradford

, Herbert

, Walker

, Home telemedicine for paediatric palliative care. Stud Health Technol Inform 2010;161:10–9

10.

Cheek

, Nikpour

, Nowlin

. Aging well with smart technology. Nurs Adm Q 2005;29:329–38

11.

Kaushal

, Barker

, Bates

. How can information technology improve patient safety and reduce medication errors in children's health care? Arch Pediatr Adolesc Med 2001;155:1002–7

12.

Williams

, Welin

, Mathews

, Towle

, Orrick

. Automated medical prescription fulfillment system having work stations for imaging, filling, and checking the dispensed drug product. See http://www.google.com/patents/US5597995 (last checked 9 January 2012)

13.

Traynor

. Diverse group meets to improve medication use in rural areas. Am J Health Syst Pharm 2009;66:1976–80

14.

Webcam comparison: Logitech QuickCam Pro 3000, Logitech QuickCam Express, Lifeview Robocam. See http://www.dansdata.com/webcams.htm (last checked 29 December 2011)