Background The operationalisation of medication reconciliation in daily practice is challenging, due to among others, insufficient interoperability of computer systems and high implementation costs. Involving patients in the medication reconciliation process by using information technology could potentially overcome these difficulties and reduce preventable patient harm accordingly.
Objective The MyMedication app (MMa) is a mobile application developed to help patients keep their medication list up to date. This study aims to investigate the usability and reliability of MMa for medication reconciliation at care transitions.
Method A random convenience sample of 17 patients was recruited during August–September 2014 at the Antonius Hospital in Sneek and Emmeloord, the Netherlands. Patients were asked to compile their medication list in MMa prior to their elective surgery. At admission, the medication list in MMa was compared with the medication list compiled by a pharmacy practitioner and the number of medication discrepancies per patient quantified. Acceptance and usability was assessed using a validated System Usability Scale (SUS) questionnaire and interviews.
Results The mean number of medication discrepancies between the medication list in MMa and the list compiled by the standard procedure was 2.5 per patient. When discrepancies caused by technical errors in the MMa were excluded, only a mean number of 1.2 discrepancies per patient remained. MMa SUS indicated good usability (68/100).
Conclusions The results of this study show that patients are positive about the use of the MMa and some feel even more involved in their medication treatment. The use of MMa for medication reconciliation at care transition shows potential as tool to improve patient safety and to reduce healthcare costs.
- Medical Apps
Statistics from Altmetric.com
Preventing medication-related patient harm, also referred as adverse drug events (ADEs), remains a top patient safety priority worldwide. Medication reconciliation is an effective strategy for preventing ADEs.1–5 Medication reconciliation is the process of creating the most accurate list possible of all medications a patient is taking—including drug name, dosage, frequency and route—and comparing that list against the physician's admission, transfer and/or discharge orders, with the goal of providing correct medications to the patient at all care transition points.6
However, the operationalisation of medication reconciliation in daily practice is challenging, due to among others, insufficient interoperability of computer systems used for prescribing in various healthcare settings. Often healthcare processes are disjointed whenever multiple caregivers are involved in prescribing medication.2 While pharmacist-led medication reconciliation process are the most effective approach in reducing unintended medication discrepancies, they may not be achievable at each transition of care for all patients at any time of the day due to time and cost restraints.3–6 Given the rising healthcare costs, pharmacist-led medication reconciliation process raises concerns regarding its sustainability in the near future. Therefore, we need to find other ways to conduct medication reconciliation at a more efficient and less costly manner.
A promising path to explore is giving the patient a more prominent role in the process of medication reconciliation by using health information technology (HIT).7–9 Heyworth et al evaluated a medication reconciliation tool delivered through a web-based portal for its value in improvement in medication safety among recently discharged patients. The authors found that this patient-mediated medication reconciliation process was feasible and well accepted.10 Lesselroth et al11 showed that ambulatory check-in kiosks, which allowed patients to review their medications before their appointment, were easy to use by patients and could be well integrated with clinical workflow. In a study by Schnipper et al, a personal medication module in an electronic patient record was evaluated on medication safety and accuracy of medication history. The authors concluded that concordance between documented and patient-reported medication lists and reduction in potentially harmful medication discrepancies can be improved by such approach.12
However, the above described patient-driven HIT tools were all designed to improve medication reconciliation process at a specific organisation (organisation bound) and are therefore prone to the same interoperability issues as those already existing with more established HIT like Computerized Physician Order Entry (CPOE) or Electronic Health Records (EHRs).13 Furthermore, use of patient portals rises confidentiality issues which could hamper their widespread use by patients. We hypothesised that a mobile application owned by the patient would empower the patient in keeping his/her medication list up to date, regardless of setting or healthcare provider, and thus limit above mentioned interoperability issues. Furthermore, since the patient would be the sole owner of the data in such a mobile application, he/she would be in charge regarding sharing the data with healthcare providers, limiting privacy concerns.
The aim of this study was to assess the reliability, usability and acceptability of a patient-operated mobile application, MyMedication app, as part of the medication reconciliation process during hospital admission.
The study was conducted at Antonius Hospital in Sneek and Emmeloord, the Netherlands. This is a 304-bed tertiary care hospital.
A random convenience sample of 17 patients of 18 years and older, scheduled for elective surgery at the Antonius Hospital in Sneek was recruited during August–September 2014.
For the purpose of this study, an already existing MyMedication app (Dutch: MijnMedicatie app; MMa) was used. MMa, is an iOS app designed and developed by the Radboud University Medical Center in Nijmegen, the Netherlands, as a first attempt to have patients create their own medication list of the medication they actually use. Screenshots are shown in figure 1. With the iPad's or iPhone's camera, the barcode on medication packages can be scanned and the information about medication name and strength matched with a medication database included in the app. When a barcode is not recognised by the MMa, patients can choose to register the information manually. Information about dose and frequency can be added manually by the patient. Furthermore, the patient can set an alarm (optional) as a reminder for taking the medication.
The MMa was offered free of charge, and the recruited patients were asked to compile their medication list in the MMa. The MMa user manual was sent by an email. At the admission to the hospital, the recruited patients were asked to email their medication list to the hospital pharmacy.
During usual care, dedicated pharmacy practitioners compiled a list of medication the patient was taking at admission. This list was compiled according to the standard operating procedure of the High 5s project of the WHO.3 The information from the community pharmacy record, the information provided by a structured interview with the patients about medication use and (if available) medication containers were combined. Such approach is considered as the most effective and preferred process in order to accomplish safe and accurate medication reconciliation at all transitions in care.3 ,14 Pharmacy practitioners are pharmacy technicians who completed a 2 years additional education primarily focused on learning clinical pharmacy skills such as medication reconciliation.
The following patient data were collected: age, gender, date of admission and type of surgery. Medication lists complied by the patients in the MMa contained medications name, dose and frequency. Medication lists compiled by pharmacy practitioners also contained medication name, dose and frequency. The pharmacy practitioner explicitly asked for the end date of the antibiotics that were used, the use of over the counter medications and medication allergies or side effects possibly caused by medications.
Patients' experiences were collected on the questionnaire forms and were actively asked for by the researcher.
The medication list in the MMa and the list compiled by pharmacy practitioners were reviewed by AWB and JA to identify medication discrepancies. The medication list compiled by pharmacy practitioners was considered the gold standard. Medication discrepancies were defined by the following criteria:
Dose discrepancy: medication listed on the pharmacy practitioner list and also in the MMa but with a different dose;
Frequency discrepancy: medication listed on the pharmacy practitioner list and also in the MMa but with a different frequency;
Route discrepancy: medication listed on the pharmacy practitioner list and also in the MMa but with a different route of administration;
Omission discrepancy: medication listed on the pharmacy practitioner list but not in the MMa and patient is using it;
Commission discrepancy: medication not listed on the pharmacy practitioner list but listed in the MMa and the patient is not using it.
The usability of the application was measured with a validated System Usability Scale (SUS) translated to Dutch.15 ,16 SUS uses a five point Likert scale and looks at three usability criteria in ISO 942-1-11: effectiveness, efficiency and user's satisfaction with the system.15 ,16 Scores can range from 0 (negative) to 100 (positive). An SUS score provides a measure of user-friendliness according to the following seven categories:
Worst imaginable: SUS score 0–12.5
Awful: SUS score >12.5–20.3
Poor: SUS score >20.3–35.7
OK: SUS score >35.7–50.9
Good: SUS score >50.9–71.4
Excellent: SUS score >71.4–85.5
Best: SUS score >85.5–100
Medical ethical approval was not necessary for this study according to Dutch trial law and the Medical ethics committee of Antonius Hospital provided a waiver for the study. All included patients gave informed consent to participate in this study.
Reliability of medication list
A total of 17 participants were recruited to this study (tables 1 and 2). The mean age of the patients was 52 years, ranging from 27 to 72 years. 59% (10/17) of the patients was male. MMa was used on an iPhone in 24% (4/17) of the patients, remainder used an iPad. The mean number of medications per patient was 3.4. Four patients used no medications at all.
Discrepancies between MMa medication history and medication list as compiled by a pharmacy practitioner included differences in dose and frequency (n=27) and differences in medications registered (n=15). The proportion of patients with one or more medication discrepancy was 64.7%. After adjustment for technical errors this proportion was reduced to 47.0%.
The total mean number of medication discrepancies was 2.5 per patient. iPhone users had no discrepancies and iPad users had 3.2 medication discrepancies per patient. When technical errors were excluded, a total mean number of 1.2 medication discrepancies per patient remained. Four patients had medication recorded in the MMa which was not identified by pharmacy practitioner according to the standard procedure. Two patients (12%) had a more accurate list in the MMa as compared with the list compiled by a pharmacy practitioner. All technical errors identified were related to the frequency button in the iPad app. Because of this error, changes in frequency could not be saved in the iPad app.
Usability and acceptance of the mobile application
The mean SUS score for the MMa was 68 out of 100. This score corresponds to good user-friendliness. Items with low scores included ‘I think that I would like to use this system frequently’, ‘I would imagine that most people would learn to use this system very quickly’, ‘I thought the system was easy to use’ and ‘I thought there was too much inconsistency in this system’.
Patient-recorded experiences and perspectives on the mobile application
Patients often reported that the MMa is a feasible tool to keep their medication list up to date and improve their medication knowledge.
‘I never knew what these medications were for until now. I feel more involved’.
However, according to recruited patients, before the MMa can be broadly implemented, several technical issues need to be fixed in order to further improve MMa usability. Also, some patients proposed new futures such as password to secure privacy of the information recorded in the MMa.
I can't change the dosing information on my iPad.
I would like this app to have a password, so other people can't see this private information.
Interviews with patients also provided valuable insight in patients' perspectives on MMa use in daily routine.
I don't think I will use this app very often, because I don't intend to end up in the hospital very often.
This pilot study demonstrated that a mobile application, designed as a tool for patients to compile an up-to-date medication list, was feasible and well accepted by the patients. The medication lists compiled by the patients were sufficiently reliable in terms of their accuracy. When compared with the gold standard (medication reconciliation by pharmacy practitioners), only a mean number of 1.2 medication discrepancies per patient were identified in the MMa, and 47% of patients had one or more medication discrepancy according to the MMa list (after correction for technical errors).
The proportion of patients with one or more (unintentional) medication discrepancies at admission varies between 60% and 67%.3 ,17 A recent multicentre study in the Netherlands showed that by implementing medication reconciliation process according to WHO procedure, this proportion can be significantly reduced to 32%.3 Based on the results of this pilot study, it can be hypothesised that the use of MMa could lower the proportion of patients with medication discrepancies by 13–20%. Since discrepancy-related ADEs are common in hospitalised patient,18 such a priori risk reduction in medication discrepancies seems very valuable.
Furthermore, by lowering the proportion of patients with medication discrepancies at admission through use of MMa, the time needed to conduct medication reconciliation and costs involved in implementing this process could be reduced. To be able to constrain healthcare costs now and in the near future, involving the patient in the process of medication reconciliation is a path worth further exploring. This pilot study has provided valuable information on how to move forward. First, the MMa should be further optimised to eliminate technical errors and to improve user-friendliness. Although, the MMa as it is (including technical errors) already scored good on SUS, the more intuitive and consistent in use the higher the chance of user satisfaction. In addition, patient concerns about issues such as data security and challenges with keeping the information updated should be taken into account.
MMa may allow patients to feel more involved in their medication treatment. Patients who feel more organised and in control of their health information by using an information technology tool have demonstrated higher intentions to use such tool.19 Therefore, for a successful implementation of MMa, patients need to feel empowered by using it.
This pilot study had a number of limitations. First, only a small sample of patients was recruited to test the MMa. The sample size was, however, sufficient to get a rich feedback on usability of the MMa in a real care setting. Second, it is likely that patients who agreed to participate in this pilot study were more skilled in using information technology and had higher health literacy. Therefore, the results on reliability of the MMa should be considered with caution. Furthermore, since we only recruited patients scheduled for elective surgery, we cannot generalise our results to other patient populations. It is important to note that recruited patient characteristics resemble the average patient population admitted to hospitals.
Further research with greater patient samples, after application upgrades, is needed. By doing so important information regarding validity of medication data entered by the patients in the MMa can be gathered as well as factors influencing data validity (by example: age, education, e-health literacy). This information is needed before tools such as MMa can be implemented at large scale. Our results show that information entered by the patients in the MMa first needs a check by a qualified person, such as a pharmacy practitioner, to assure that the medication list is accurate and complete. Therefore, we strongly feel that when used on large scale, the MMa will have the most value when used embedded in the process of medication reconciliation. Healthcare providers who perform medication reconciliation on admission to the hospital, in the Netherlands usually pharmacy technicians or practitioners, will have a more accurate list to start with and the patient will feel more empowered in understanding and monitoring his/her pharmacotherapy. Also, the cost-effectiveness of the MMa at different care transitions should be taken into account when exploring the use of this app on a large scale. For example, the process of obtaining an accurate and up-to-date medication list for patients who visit an outpatient clinic is usually not pharmacy led in the Netherlands, due to limited staffing. Therefore, the need for other approaches is probably far greater in that particular setting in comparison with hospital admissions.
Furthermore, in the Netherlands, patients can give consent to share their current and historical medication data through the so called National Health Exchange Platform (Landelijk Schakel Punt). This platform connects hospitals, general practices, outpatient and home pharmacies with each other. This enables healthcare providers involved in the care of the patient to exchange medication information in a fast and reliable manner. If interoperability of the MMa with this platform could be achieved, the medication information already available in the platform could be imported into the MMa and enriched by the patients using the MMa. By doing so, the need to enter all medication data manually by the patient would be reduced and thereby lowering the chance of making errors during that process and improving user-friendliness.
In addition, the perspectives of physicians on MMa use, as medication reconciliation tool, were not studied as well as the patient–MMa–physician interaction during a consultation. These aspects should be considered in future studies, as the role of the organisation and healthcare workers in empowering and stimulating patients to use information technology is a very important influencing factor to enhance use of tools.19
Involving patients in the medication reconciliation process by a mobile application such as MMa shows great potential to overcome barriers in implementation of this process at care transition. However, further development of MMa and research is needed to gain insight in how to successfully implement such an application. Also, further research is needed to investigate which types of patients use an application such as MMa safely and efficiently in order to embed MMa in the medication reconciliation process.
The authors thank all staff of preoperative intake ward and hospital pharmacy department of the Antonius Hospital in Sneek, the Netherlands for their support, and in particular, REshape Innovation Center for building the mobile application for the study. They would like to thank Paulina Stehlik of the University of Sydney for doing a language edit.
Contributors JEK, MD, LJLPGE and JA conceived and designed the experiments. AWB and JA performed the experiments. AWB, JEK and JA analysed the data. AWB, JEK and JA wrote the paper. LJLPGE and MD critically reviewed the manuscript.
Competing interests None declared.
Patient consent Obtained.
Ethics approval The Medical ethics committee of Antonius Hospital provided a waiver for the study.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.