Nursing home staff’s experiences with the implementation of an automated blister packaging system for solid, orally ingestible drugs: a qualitative study in Northern Italy

Abstract

Objectives

To explore nursing home (NH) staff’s experiences implementing an automated blister packaging system for solid, orally ingestible drugs.

Design

This was a descriptive qualitative study. Semistructured interviews were used for data collection, and qualitative content analysis was applied for data analysis.

Setting

Four NHs in a region in Northern Italy participated in the piloting of an automated blister packaging system.

Participants

The purposive sample comprised 40 NH staff (seven registered nurses, 11 certified nursing assistants, three nursing aids, seven nursing managers, four directors and eight physicians)

Results

Six main categories emerged from the data analysis: (1) impacting safety, including improved medication process, perceived insecurities and remaining error sources; (2) creating free spaces, including time, individualised care and drug storage and logistics; (3) serving residents, (4) meeting expectations, including met and unmet expectations; (5) generating efforts and (6) producing waste. They all merged into one theme ‘It pays off!’

Conclusions

As experienced by NH staff, the automated blister packaging system improved medication safety and reduced the workload of registered nurses, but it is not an ‘all-around carefree package’. Continuous quality improvement and risk management strategies are recommended to accompany and sustain implementation, as well as further clarification of roles and duties among the different care workers involved in medication management. Further research is needed to gain a better understanding of the impact of an automated blister packaging system on registered nurses’ competencies in NHs regarding medication management and residents’ therapies and safety.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Applying a qualitative approach to this under-researched topic revealed new aspects regarding the implementation of an automated blister packaging system.

• The strength of this study is that data were collected by seeking perspectives from various points of view, for example, from nursing home (NH) directors to nursing aids, by exploring their perceptions and experiences.

• The extent to which the four volunteering pilot NHs are representative of all NHs in this Italian province remains unclear, as they might be characterised by a high willingness to innovate. This factor must be considered when scaling up automated blister packaging systems for other NHs.

• As the new automated blister packaging system was implemented in the four NHs with mainly positive experiences despite the differences in structure (eg, size, staffing levels and mix) and the adaptation approach, the transferability of the results to other NHs is encouraging.

Introduction

Due to the demographic change as well as higher incidence of chronic diseases, improved medical care and changes in family structure the number of older multimorbid adults in need of long-term care, including nursing home (NH) care is increasing worldwide. NH residents often take a high number of medications and are at risk for side effects, adverse drug interactions, medication errors and adverse drug events.¹ Side effects can occur due to the altered pharmacodynamics and pharmacokinetics in old age (eg, stronger effects of medication compared with younger adults).² Medication management in NHs is complex and medication errors are common. According to international studies, 16%–27% of NH residents are affected by medication errors, such as the administration of wrong medications or incorrect dosages, which can have serious yet infrequent consequences for NH residents, including hospitalisation or mortality.³ Errors can be made by all professionals involved in medication management (eg, care workers, physicians and pharmacists), including prescribing, dispensing and medication administration errors.⁴ Within medication management, the manual ‘preparation’ of medications by the nursing staff is a time-consuming and error-prone process (eg, due to misleading and confusing labelling of blisters, work interruptions during manual preparation activities and as a result mix-ups of medications).

The use of technologies, such as automated blister packaging machines, is a promising measure to reduce medication errors and increase resident safety.^{5 6} Automated blister packaging can be defined as the automated repackaging of finished medicinal products in a non-reusable container individualised for NH residents and their specific dosing times.⁶ The service of unit dose repackaging of medications for oral solid dosage in a disposable container (blister pack) according to a person’s medication regimen is often offered by blister packaging centres within pharmacies. In some countries, the implementation of these systems is strongly promoted by health policy, arguing for improved safety and the saving of resources by outsourcing the blistering of medication to machines/robots.⁷ Considering a shortage of sufficiently qualified care workers, the time spent manually blistering of oral unit dose solids could be used for other, more direct care activities for residents. Most existing studies on automated medication blister packaging systems were conducted in the hospital setting. A recent systematic literature review found that the main benefits following the implementation of an automated blister packaging system are the reduction of medication errors, consequently improved patient safety, time savings in medication preparation and dispensing, as well as cost savings through reduced inventory, labour savings and waste.⁷ However, the implementation process is often complex and time-consuming, and the cost–benefit ratio has not yet been demonstrated due to the lack of controlled studies in this field of research. Introducing such a system changes work processes and daily routines in medication management, including drug prescription, dispensing/blistering and administration. A non-optimised process and lack of reliability of the automated blister packaging system can lead to inefficient work and dissatisfied staff.⁵ Craswell et al evaluated the implementation of an automated blister packaging system in an Australian hospital. They found that the new system had an impact on nurses’ workflows leading to workarounds/deviations from work procedures, which in turn might negatively affect patient safety.⁸

Although centralised automatic medication blister packaging in NHs is expected to have similar benefits to those in hospitals, such as reducing the workload and saving time for care workers when preparing medications, there is currently no evidence to support this.⁶ To date, few studies have investigated the experiences of implementing an automated blister packaging system from the perspective of the staff involved, particularly in NHs. In a northern Italian region, within a pilot project promoted by the regional Nursing Home Association and a private pharmacy, a centralised automated medication blister packaging system was implemented in four NHs between March and August 2022. This study aimed to evaluate the implementation of this centralised automated medication blister packaging system by exploring the following research questions:

• What are the experiences of the NH staff (management, care workers and general practitioners) in the four pilot NHs in medication management with the centralised automated blister packaging system?

• What effects on ‘daily care’ (eg, resident care, work organisation and processes and time resources) do care workers experience due to the implementation of the centralised automated blister packaging system?

Methods

Study design

We used a qualitative descriptive research design (embedded in a naturalistic inquiry) to explore NH staff’s experiences with the implementation of the centralised automated blister packaging system on medication management, as well as its impact on ‘day-to-day care’. Qualitative descriptive research ‘is the method of choice when straight descriptions of phenomena are desired’ (p. 339).⁹ We employed qualitative content analysis for data analysis according to Kuckartz and Rädiker (2022).¹⁰ The Standards for Reporting Qualitative Research (SRQR) were used for reporting this study.¹¹

Setting

We conducted the study at four NHs in South Tyrol, a province of Northern Italy, 6–12 months after the implementation of the automated blister packaging system. Overall, there are 79 NHs in the province, which by law are considered social care institutions. The largest NH that participated in the study had 138 beds, and the smallest 46 beds.

South Tyrolean NHs, including the participating pilot NHs, employ a mix of different care workers. Registered nurses (RN) in Italy, since 1999, must complete academic nursing training and qualifying with a bachelor’s degree. Certified nursing assistants (Sozialbetreuer/operatore socio-assistenziale) (CNA) complete a 3 year vocational training in social and healthcare and work in NHs or other social settings. Nursing aids (NA) in general have a 1 year training in basic care and work in the social or healthcare sector.

RN have overall responsibility for the medication process in the NH; specifically for ordering medicines; storing and monitoring drugs; for manual blistering (dispensing) of medicines to prepare the blistered/dispensed medicines for daily distribution; handling changes in case of therapy change; for administering or supervising the administration of medicines to the residents; for monitoring the intake of and efficacy of medicines and for communicating effectively with the physician. Certified nurse assistants, under the supervision of the RN, share responsibility for distributing and administering medications and for communicating and reporting to the RN the effects of medications they observe in the resident. They also help in the daily preparation of the blistered medications for distribution. NA, under the supervision of the RN and the CNA, are allowed to distribute and administer medications.

Detailed information is presented in table 1. All four NHs have chosen voluntarily to participate in piloting the centralised automated blister packaging system.

Table 1. Characteristics of the participating NHs (reporting date: 31 December 2022).

Characteristics	NH 1	NH 2	NH 3	NH 4
Number of beds	46 (no short-time care beds)	96 (5 short-time care beds)	138 (8 short-term care beds)	90 (5 short-time care beds)
Number of residents	26	96	110	90
Number of nursing wards/areas	2 (1 area closed)	3	7 (1 area closed)	3
Number of physicians/general practitioners	2	5	3	3
Number of care workers in full time equivalent (RN, CNA and NA)	RN: 3.3CNA: 9.35NA: 6.05	RN: 13.69CNA: 23.66NA: 24.69	RN: 10.50CNA: 17.75NA: 40.85	RN: 8.57CNA: 16.26NA: 19.87
Number of NM	2	1	2	1
Number of directors	1	1	1	1

CNAcertified nursing assistantsNAnursing aidsNHnursing homeNMnursing managersPHphysiciansRNregistered nurses

Description of the automated blister packaging system

The pilot project for implementing the automated blister packaging system in the four NHs was designed, planned and introduced by the provincial Nursing Homes Association (https://www.vds-suedtirol.it/), in close collaboration with a private pharmacy in South Tyrol (Italy), which offered this service. The Nursing Homes Association and the blister (packaging) centre within a private pharmacy supported the NHs in implementing the automated blister packaging system (SiCuro^®). The automated system comprises software and hardware and was implemented in this pilot project by the pharmacy for deblistering and automated production (blistering) of solid, orally ingestible drugs as individually packed patient/resident blister bags. Implementing the automated blister packaging system led to a reorganisation of the medication management process within the four pilot NHs, as illustrated in figure 1.

Figure 1. New medication management process after the implementation of the automated blistering of oral medication.

The phases of the new medication management process consist in the following:

1. Prescribing of medication by the residents’ physicians in the new (specific) software.

2. Ordering of the medicines of the NHs for the blister centre by RN utilising the specific software: medications must be ordered in the public hospital pharmacy (as regulated by Italian law).

3. Pick up (collection) of the ordered and original packed medications in the hospital pharmacy and transport/delivery to the blister centre by the NHs.

4. Receipt of medications in the blister centre (each medication pack will be labelled with a batch and serial number for traceability right down to the individually blistered tablet)

5. Production, control and approval of the blisters: it is a blister packaging system for solid, oral forms of medication that produces both single doses and multiple doses as tubular blister packs in chronological order individually for each patient (for 1 or 2 weeks). Each blister pack contains information on the patient, medicine (trade name and active ingredient), administration time and a barcode with a serial number. For the production approval process, each blister pack produced is subjected to a visual inspection by the ‘Pouch Inspector’ system. Once all checks have been completed, the production is approved and electronically signed by a pharmacist when the batch report is created in the specific software.

6. Delivery of the blister bags (rolls) from the pharmacy/blister centre to the NHs.

7. Daily preparation and distribution of the blister bags to the individual residents at the prescribed time. The supplied blister bag rolls are stored centrally in the medication room or the single nursing areas of the NH, most commonly in a lockable medication trolley. At the time of distribution, nursing staff tear the single blister bags directly from the blister roll and administer it to the resident.

8. Documentation of administered drugs in the nursing software (SENSO^® 7).

9. Monitoring of intake and efficacy of therapy.

The main difference between the previous and the new system is that the ‘blistering activity’ is now outsourced to a private pharmacy, which substitutes the manual preparation done by nurses within the NHs. Before, oral medications were then manually blistered/dispensed, usually once a week for all residents and for a whole week. With the new system, these medications are automatically blistered/dispensed in the blister centre within the private pharmacy for 1 or 2 weeks. Still the medications need to be checked and prepared daily for distribution/administration to NH residents.

While before the implementation of the automated blister packaging system physicians prescribed and updated the pharmacological therapy for every resident utilising the nursing documentation system (SENSO^® seven software), a new software has been introduced for prescribing the medications (SiCuro). With this new software, the ordering of the medications from the public hospital pharmacy has been simplified for nurses. While before medications were stored within the NHs (for manual blistering), they are now delivered to and stored at the blister centre. Instead of the originally packed drugs stored in medicine cabinets, NHs receive packaged rolls for each NH resident from the blister centre. The administration of drugs was and still is documented in the nursing documentation system SENSO^® 7 (software).

Participants

To explore the experiences with the implementation and utilisation of a centralised blister packaging system, the participants in this study were purposefully sampled from the NH staff of the four piloting NHs. NH staff comprised personnel directly involved in the medication process, such as care workers and physicians, as well as management, such as nursing managers and directors of the NH (see table 1). Participants were included voluntarily and when they had been affected by the new automated system (ie, they worked already in the NH before the pilot took place).

A total of 40 participants were recruited during the interview process. The sample comprises seven RN, 11 CNA, three NA, seven nursing managers (two of them were vice nursing managers), four directors and eight responsible physicians, who in most cases were appointed general practitioners. All of the participants had gathered experience with the new blistering system during the pilot project.

Data collection

For organising the interviews, we called each NH individually and talked to the NH director and/or nurse manager. They had already received the first information about the evaluation study through the Association of Nursing Homes. We explained in more detail the purpose and methods of the study to them, whom we would like to interview, and how to conduct the interviews (single or in groups per NH: RN single or in a group; other care workers in a group; physicians of the same NH single or in a group; nursing managers single and NH directors single). The nursing manager and director of the NH received all written information material (information leaflet about the study, information leaflet about data protection and informed consent document) and were asked to distribute it to all personnel so that everybody had a chance to know about the study and to take part if wanted. The nursing manager organised the day on which the researchers could come to interview the participants, as well as the time slots for the interviews on that day.

Data were collected through face-to-face, semistructured interviews on 4 days (1 day in each NH) during March and April 2023. In total, 22 interviews, nine single and 13 group interviews were conducted by the three researchers (WT, HW and DA), after which we agreed that the collected data were sufficient to understand and describe the patterns and themes and no additional interviews were needed. In most of the interviews, at least two of the researchers were present, one conducted the interview, and the other one took notes or asked questions towards the end, to complete the interview. An interview guide was used as outlined in table 2. The interviewees could talk either in Italian or German, which are the main official languages spoken in the region. Interviews lasted between 30 and 45 min. They were audio recorded and transcribed verbatim. The transcribed documents were checked (by WT, HW and DA) for overlooked names which could have indicated the identity of the NH or a person, which were then deleted so that no identifying hint appeared in the text. Each interview document received a pseudonymised code.

Table 2. Interview guide.

No	Key questions
1	How does the new medication process work?
2	How did it work before? What are the differences between the two systems?
3	What expectations did you have before the pilot project started?
4	What has changed in the nursing home?
5	What are the benefits, advantages and disadvantages of the new blistering system?
6	How has the automated blistering system affected the time resources of care workers as a whole or of the individual professional groups (eg, for nurses)?

All participants received written information letters about the study and data protection before the interview. They were also informed orally about the study and how confidentiality and privacy/data safety would be guaranteed directly before the interview. In case of uncertainties or unclear points, they were invited to ask questions; they were also told that they could withdraw at any time from the study without consequences. Participants volunteered to be interviewed, and audio recorded and provided informed consent before starting the interviews.

Trustworthiness during data collection

The researchers were not involved in the design and implementation of the automated blister packaging system. To get a better understanding and insight into how the new system works, the researchers read the project description thoroughly and visited the private pharmacy and its blister centre, before approaching the NHs and collecting data. Two of the researchers are PhD-prepared RN (WT and DA), one with previous work experience in a NH. One has done her doctoral thesis within the NH context (WT). The third PhD-prepared researcher has a background as a nutritional scientist with several years of experience as a researcher in the healthcare sector (HW). None of the researchers had direct contact with the pilot sites before the study.

Additionally, after the interviews, the research team was invited by the local staff/nurse managers to take a walk round with them in the NH and in the wards where the new blister packaging system was working. They explained how it works and pointed at the practical side of the new medication process, the new software and equipment they used, etc. During these activities, the research team took field notes about the functioning and observed variations within and between NHs.

Debriefing meetings between the researchers (WT, HW and DA) took also place after each interview day, in which all interviews of one specific NH were carried out. Observed particularities, differences, or commonalities were written down as notes.

Data analysis

Data analysis was performed using qualitative content analysis according to Kuckartz and Rädiker,¹⁰ using the computer-assisted qualitative analysis software MAXQDA Analytics Pro 2022 (VERBI Software GmbH), in which the transcribed and pseudonymised text documents, as well as the audio files, were imported. The analytical process consisted of five steps carried out by three researchers (WT, HW and DA), experienced in qualitative analysis:

Step 1: getting acquainted with the textual data: each researcher thoroughly reviewed the transcripts and marked the text with relevant content (the text documents were divided among the three researchers); initial memos were written.

Step 2: creating main categories based on the interview guide: initially, the marked (highlighted) text passages were coded deductively into the main categories derived from the research questions and the interview guide. During this step also, initial subcategories per main category were created inductively, out of the material. Memos were written.

Step 3: refining categories and subcategories and providing definitions: steps 1 and 2 were performed by all three researchers independently; each researcher coded one interview individually, after which we compared and discussed our categories, emerging subcategories and coded text passages in a combined session. Through an iterative process, we generated an initial category system. Simultaneously, definitions for the main categories and subcategories were written. According to Kuckartz and Rädiker, definitions should be as precise as possible so that the research team has the same understanding of the categories.¹⁰ In that way, each researcher had a shared and agreed-upon guide (‘codebook’) for the coding process which was also thought/intended to enhance dependability.

Step 4: coding and continuous adaptation of the category system: all remaining transcript files were divided among the researchers, and all material was coded (independently) applying the newly created category system. In several meetings, the researchers shared their analytical activities and outcomes to discuss, reflect on and further refine the category system to ensure that all researchers applied the coding process consistently and transparently. The researchers encouraged each other to express contrasting views or understandings. In case of discrepancies, it was discussed till an agreement was reached. The use of memos and field notes was helpful in this ‘clarifying’ process.

Step 5: reporting results: in the final step, the researchers interpreted and refined their analysis and order of the main categories and subcategories by writing the study report. Also, this final analytical step was done in a shared way.

Study rigour—trustworthiness

To assure and enhance the quality of the present study, the reflections highlighted by Polit and Beck about the Lincoln and Guba framework of trustworthiness were considered.¹² Credibility and dependability were enhanced by including an adequate number and a wide variety of participants, and all three authors participating and performing data collection and analysis. Making analysis transparent and discussing both findings and discrepancies and differences of interpretations/understandings between all three researchers regularly until a consensus was reached helped increase confirmability. Additionally, the final draft of the report was handed out to the responsible staff (RO and JP) of the pilot project (Nursing Home Association) to check its plausibility, thus adding to the confirmability of the results. Dependability was enhanced as more than one researcher participated in data analysis and by following a shared ‘codebook’/coding guide. The description of participants/setting characteristics, data collection and analysis should allow readers to judge the transferability of the study findings to other contexts.

Patient and public involvement

No patients/residents were involved in this study.

Ethics approval

Participation was voluntary. Before each interview, participants were provided with an oral explanation of the study. They were also informed that they could withdraw from the study at any time. Throughout the research process, participants’ privacy was protected using a numerical identifier instead of names. NHs were referred to by using a number. Confidentiality of the data was strictly maintained. Audio recordings were destroyed after transcription, while transcripts will be securely stored for 5 years before being destroyed.

Results

Through the analysis of the data regarding the experiences of care workers, nurse managers, physicians and NH directors with the implementation of the automated blister packaging system, we identified six main categories, (1) impacting safety, including the subcategories ‘improved medication process’, ‘perceived uncertainties/insecurities’ and ‘remaining error sources’; (2) creating free spaces, including ‘time’, ‘individualised care’ and ‘drug storage and logistics’; (3) serving residents; (4) meeting expectations, including ‘met’ and ‘unmet expectations’; (5) generating efforts and (6) producing waste. They all merged into one theme ‘It pays off!’ (illustrated in figure 2).

Figure 2. Category system (overarching theme, categories and subcategories).

Category 1: impacting safety

Ensuring the greatest possible safety for residents when administering drug therapy is the priority to avoid unintentional errors in prescribing, preparing and administering medication. The interview participants from all four pilot NHs described that the implementation of the automated blister packaging system had a considerable impact on increasing medication safety, for example, through SiCuro software, clarification of tasks within the care worker team, clear and hygienic blister packs, but that perceived uncertainties and potential sources of error remained even with the ‘new system’.

Subcategory 1.1: improved medication process

Compared with the previous (manual) blistering process, automated blister packaging has reduced potential sources of error, particularly in the preparation and administration process (ie, preparation of single-dose medications for each resident) as most interviewees confirmed. Care workers felt safer overall: ‘And as I said, administering medication is much safer for all nursing staff, not just for us nurses … you now have little blister bags, and everything is in them (and written on them)’ (NH3-RN). One physician in charge highlighted the positive impact of the new system: ‘It’s not like before when mistakes were made … but now it (the blister bag) comes with the person’s name, surname and room number, so you can tell whose it is: I haven’t seen any mistakes, honestly, since this method was introduced’ (NH4-PH2). Drugs cannot bounce out of the bags, as they are sealed, and this reduces mistakes. With the manually prepared blister boxes, for example, if they fall, nurses had to check which drug had leaked out, replace it, and often the blister box (divided into three sections for the daily medication: morning, midday and evening) had to be prepared anew.

The new SiCuro software played an important role in this improvement as it works as a reliable and updated source of information for care workers regarding therapy plans and drugs themselves. CNA found it helpful: ‘Yes, and if I didn’t know it, then I go to the computer and look at it and know exactly which drug, or which colour or which shape, which label it has. And that suits me really well! Because it gives me certainty about what I'm administering’ (NH1-CNA2). Nurses appreciated that actualised therapy plans are now in the system, as ‘with SiCuro® it is the case that the doctor must look at the therapy, see what has changed, and flag it off. Otherwise, the prescription/therapy won’t be released’ (NH3-RN).

With implementing the new system, it was also necessary to (re)clarify the tasks and roles of care workers involved in the medication process, particularly those of nurses and CNA. This process was not yet complete by the time the study took place. ‘Every nursing home does it a little differently’ (NH2-NM), as a nursing manager said, partly due to the staffing situation. While in some NHs, the preparation of the (new) blister-packed medication for distribution to the residents was carried out by CNA, in other NHs, this task was carried out exclusively by RN: ‘[…] We have said that as long as we have enough nurses, we want to do this (preparing blistered medications daily for distribution) with nurses only. And that means there is a nurse in the medication room every day between one and three o’clock in the afternoon’ (NH2-NM).

The automated blister bags were described as clearer and more hygienic than the old blister boxes: ‘I am now also of the opinion that it is better in terms of hygiene’ (NH3-NA). The blister bags ‘have everything nicely labelled’ (name of the person, room number, medication, dosage and time of administration) and the information is easy to read: ‘no longer in the doctor’s or nurse’s handwriting … that is an advantage’ (NH2-CNA). If necessary, medication can be crushed in the blister bag.

Subcategory 1.2: remaining error sources

Despite the positive impact on medication safety, participants pointed also to some potential error sources remaining even with the ‘new system’, for example, when distributing and administering medication. A CNA made it explicit: ‘And in the end, with the system, we are much more protected than before. Mistakes can only happen, but it has always happened, that I give a resident the wrong blister bag. But then that’s my fault’ (NH4-CNA).

Particularly underscored as a new source of error was the therapy changes between the blistering cycle (at the time of the interviews). Blistering cycles were 1 or 2 weeks, meaning that the therapy of a resident was either prepared for a whole week or even for 2 weeks in the blister centre. The NH could choose between these two options. So, when a change of medication was necessary for a resident after the NH had received the prepared blister bags for 7 or 14 days, nurses had to remove manually the discontinued medication from every single blister bag and add manually the newly prescribed drug. This involves cutting open the blister bag and resealing it: ‘I see this as a major disadvantage with the therapy changes because that’s where it gets complicated and that’s a source of error’ (NH3-PH).

Caution and prudence are still needed in the medication process, as not the entire process is automated. The new system does not eliminate all possible sources of error: ‘It’s okay the way things are going. But it is perhaps an illusion to believe that everything is safer with it. People still must watch out’ (NH4-PH2). A nurse manager for example had observed how some mistakes were made: ‘But it has already happened two or three times that they have distributed the drugs at the wrong time because they have not read what’s (written) on the bag. The bag was all accurate with name and so on […], so there are still a few small sources of error, but on the human part, from us, and not from the machine’ (NH2-NM).

Subcategory 1.3: perceived uncertainties/insecurities

Alongside the predominant benefits, RN expressed insecurities and uncertainties. Some pointed at the perceived loss of knowledge about residents’ therapies and medication in general as the automated blister packaging system changed the way medication was administered.

‘And another disadvantage from my point of view is that the therapies, for example, when I used to prepare/blister the medication, I also looked at the therapy a bit critically. Yes, this resident has been on two diuretics for a fortnight, perhaps it would be better for me to ask the doctor whether we should continue or not!? Now that everything is automated, I can no longer look at the whole therapy every day. And that’s a bit difficult or should be reconsidered because the nurses also make a lot of suggestions for the therapies’. (NH1-RN2)

This perceived drawback highlights the important role and responsibilities nurses have, concerning the whole medication process in NHs; for example, highlighting criticalities about residents’ therapy plans to the physician who usually visits their residents once a week to check whether certain drugs should be continued or discontinued, or something else is needed.

Giving up old working habits, or even just single tasks is perceived as a disadvantage by some nurses, as it appears to them like losing the overview. One nurse meant that ‘If you no longer prepare the drugs (manually), you must take extra time to look at the therapy (plan) so that you can see what you hand out in the morning, what you hand out at lunchtime, (although) the blister bags show exactly what the resident gets. But you only have an overview when you look at the therapy (plan)’ (NH1-RN1). Manual blistering seemed to have the benefit that the residents’ therapy plans were memorised by nurses, which provided them with knowledge, available at any time (that can be called up), without having to check the charts.

Feelings of uncertainty were also expressed about losing a part of their area of competence/responsibility. Nurses felt that activities regarding medication are one of their key professional aspects and change can be threatening. ‘Now this expertise is also being taken away from us’ (NH1-NM). This statement points simultaneously to another issue, namely the latent message that even other (clinical) activities and competencies of nurses have been lost in the specific setting of the NH. However, the nursing manager, who gave the above example of a nurse’s reaction and feelings of insecurity, countered that the expertise of nurses is not just about medication management. ‘It is about much more’, she said. Defining well, which professional group does what in the medication process, as already highlighted earlier is an important aspect for nurses; in some NHs, it has been clarified but in others not yet, there is still a bit of ‘confusion’; it is important not to have a ‘grey area’ in this respect, but ‘to define that (roles and responsibilities) well. That makes the difference’ (NH1-RN2).

Category 2: creating free spaces

Based on their experiences, the study participants described the ‘free(er) spaces’ created by implementing the automated blister packaging system in terms of time, individual care, and medication logistics.

Subcategory 3.1: time

Nurses, in particular, found that some time has been freed up. Time which could be saved due to the elimination of manual blistering has been estimated by some to be about 2 hours daily. However, the perception of this ‘free time’ varied between NHs depending on the staffing situation, and how the medication process was adapted to the new system. For example, some reported less strain to always prepare the weekly medication on time: ‘And the pressure was certainly taken off. Because having to prepare the drugs and do all the organizational stuff on the day of the ward round (with the physician) was a lot of pressure’ (NH3-NM). CNA agreed with nurses on this point: ‘At most. I think it’s simply about relieving the burden on the nurses. I also think that the system will certainly have come about because there are few nurses’ (NH3-CNA3).

In one home, the time saved meant that the working hours of nurses were reduced (due to the shortage of nurses) or reallocated to CNA. This means that certain tasks within the medication process have to be reorganised, and even shifted to CNA: ‘But it’s also a good thing that the assistants can do this (eg, preparing non-blisterable drugs like drops, sirup), and you have to know more, because in the future, as it is now, without nurses, you have to get someone for these things (handling medications)’ (NH1-RN2).

One NH director summarised that ‘the new medication process allows to free up time’ but is aware that the system cannot run ‘nurse-independent’, and further ‘So even if you have fewer nurses, you guarantee an excellent service in terms of medication preparation’ (NH2-Direktor). For NH directors, the new automated blister packaging system is a way of ensuring the quality of services and safety, even in times of a shortage of RN, and by redistributing tasks and responsibilities. They also highlighted that it makes more sense to automate certain tasks, to free up time that can be used directly for residents’ care. A resident’s physician emphasised in this regard that: ‘It was easier for us before. But I think it’s more of an advantage in general, for the nursing home, for care. Because if the carers simply have time with the people, with the residents, and are not simply there doing things that can also be automated’ (NH3-PH1).

Subcategory 3.2: individualised care

Automated blistering (and the time saved by not having to do it manually) frees up space for resident care. The residents benefit directly ‘The ‘saved’ time is spent now anyway for the resident’ (NH4-RN1).

In one NH, the time freed up by the new system got transferred to CNA who now dedicate ‘two hours explicitly […] for the residents, for (creating) moments of happiness, for moments of well-being, that they simply enjoy. From feel-good baths to colouring their hair once or going to the Christmas market…’ (NH1-CNA2).

Nurses mentioned also that now they have some space for indirect care tasks like: ‘… care planning has fallen by the wayside. […] and now you can take time for these things again. With the two or three hours you have now’ (NH1-RN2). These gained hours allowed nurses also end the shift timely: ‘And at least you don’t work overtime for these things (care planning, organizational work). Because going home on time is also a nice thing’ (NH3-RN).

Subcategory 3.3: medication storage/logistics

In addition, free space has also been created regarding the medicines stored in the ward. There is now significantly less medication in stock and less space is required for this.

‘And now we have fewer drugs in the nursing home. Because now we only have basics like Lasix, Plasil, or Novalgin, things you need sometimes… but otherwise we had three cupboards full of medication’ (NH4-RN)

It takes less time to store the medicines and check the stocks (eg, expiry date) and the lower stock levels also mean a better overview and control: ‘What is no longer needed … putting away/clearing the ordered medication and distributing it to the nursing wards, the night shift usually did that, and it was also very time-consuming’ (NH2-NM). In this sense, the new system brought about lean work procedures.

Category 3: serving residents

Initially, care workers had concerns about handling, but ‘independent’ residents can open the new blister pack autonomously and take the medication on their own. ‘So now I have the bag and everything is tip-top inside and the residents also report that it is very easy to open it themselves. And you open it a bit for the others and then they do the rest themselves’ (NH4_CNA).

Blister bags are practical even if residents want to take the medication into the dining room or their room. In addition, it is now easier to track if a resident has not taken their medication as the medication is ‘lying around’. ‘Because (with the old system), if he/she leaves it somewhere or forgets it, you no longer know who it’s from’. And with the new blister bags that does not happen anymore: ‘That’s just good’ (NH3-NM). Also, relatives appreciate the new system. One nursing manager underscored that serving residents with the new system ‘[…] is a good feeling. So that speaks of quality, care improvement and safety. And that has also gone down very well with the relatives’ (NH4-NM). There was positive feedback from the relatives about the blister bags when they went on an excursion or holiday together. For example, one NH director stated: ‘… a resident was picked up by her family because they were going on holiday […]. And we were able to give the relatives the blister-packed medication for the following two weeks. And at a family meeting the family said: Wow, that worked so well, it was fantastic’ (NH2-Director)

Category 4: meeting expectations

Study participants reported being overall satisfied and that their expectations regarding the new system had largely been met, as demonstrated within the previous categories (eg, improvement in medication safety, reduction in workload). The introduction of the automated blister packaging system initially caused a ‘huge fuss’, but ‘in the end, it has more advantages’ (NH2-CNA3), ‘it’s paying off’, so the overall opinion. At the same time, some unfulfilled expectations (eg, major time savings) were also described, which is why a few respondents were neutral about the new system or would like to ‘go back to the old system’ if the opportunity arose, as one interviewee expressed.

Subcategory 4.1: met expectations

Most respondents were very satisfied and would recommend the new system to others because the advantages outweigh the disadvantages: ‘And then also as coordinator, I have the feedback from my colleagues, and they are all very happy. Apart from an initial moment of course there was a bit of an adjustment to do to understand the various mechanisms well, but once you get into the routine, it goes quite well’ (NH4-RN)

The new system offered several simplifications and facilitations. From the point of view of a residents’ physicians, for example, the SiCuro software made it possible to prescribe therapy from home: ‘… you can also use the programme from home. If you have an enquiry on the phone, you can also enter the prescription from home if necessary. That’s very convenient’ (NH3-PH2). Nurses particularly welcomed that ‘ordering medication (from the hospital pharmacy) is now almost easier to do, simply because we can order everything we need in SiCuro’ (NH3-RN).

Nurses perceived that their expectations have been met, especially as the workload has decreased: ‘And my personal experience is also, because I, we did manually blistering before and prepared the drugs in the weekly pill boxes, and when you have to prepare medication for 30 residents for seven days in six to seven hours, we both had this workload, […] and also felt its side effects and therefore it is a very, very big advantage to switch (to this new system)’ (NH4-NM). The experiences towards the new blistering system can be summarised with the quotation of a nurse: ‘In the end, expectations have already been met and things are going almost better than expected. It seems to me. That’s great’ (NH3-RN)

Subcategory 4.2: unmet expectations

Staff, however, also expressed some unmet expectations and were neutral about the new blister packaging system: ‘My job has not changed, honestly because I always do the same, I prescribe and then it is administered’ (NH4-PH3). Thus, despite the benefits it has for residents’ medication safety, the new system did not change or make work easier for physicians.

A director of one NH added a critical aspect from the project management perspective: ‘A lot has been promised: You need less time; everything is easier and so on. But every change at the beginning is wavelike’. The introduction of technological innovations had an impact on work processes and entailed changes for the employees concerned: ‘That’s why you must invest time when you start a new project. And the feedback in the first three months was not so positive’ (NH2-Director).

What one CNA experienced as positive ‘No, I wouldn't go back. It’s well-established now and that’s great. There are a few subtleties that need to be improved but in general. […] So it suits me fine’ (NH2-CNA3), for another one it was not, he/she could not see an advantage in the new system: ‘Yes, but I think the old system was almost better, from my point of view. In terms of handling for sure, 100 per cent’ (NH3-NA). The handling here refers to the automated blister bags (tearing them open when assisting the resident, etc).

Category 5: generating efforts

As with any change in routine, the implementation of the automated blistering generated several additional ‘efforts’. Automated blistering helps to save time, but it cannot be applied for short-term care residents (at the time the study took place). This means that the medication to be taken orally for these residents still must be blistered manually: ‘… because we don’t use the new blistering system for transitional care, for those residents who come to us for a short time, because we don’t know how long the resident will stay and many therapies often change during transitional care. That’s why we’ve stuck with the old preparation’ (NH1-RN2). Likewise, preparing non-blisterable drugs (orodispersible tablets, liquid medication in oral drop or syrup form) must be done manually as it was before.

But also, therapy changes during the blistering cycle required some extra effort from nurses, as they have to take out discontinued drugs from and/or newly prescribed drugs into the blister bag manually, like a CNA described: ‘I think they are blister-packed 10 or 14 days in advance. And then logically, when it arrives and after you have a change of therapy, you must tear it open. All the bags (of this resident)! Put the new drug in and tape it up again. It’s a lot of effort, you must admit. (NH2_CNA3). Also, a resident’s physician underscored this extra work as being ‘a bit cumbersome for the nursing staff, … if someone has something acute, for example, heart failure increases, you need more diuretics, you prescribe it, and then they must add it (to the already prepared blister bags) and it only gets (automatedly) blister-packed a week later’ (NH2-PH1). One nurse pointed at a specific clinical situation of persons with cognitive impairment, ‘when people are very restless in the dementia area’, and when the right therapy and dosage must be found: ‘… you also have a lot of changes because you must keep adjusting the therapy. And that’s an effort in time required for preparation afterwards […]. If you have residents who are well and stable, then there are hardly any changes. But there are phases’ (NH2-RN)

The new system potentially requires nurses and physicians to discuss the need for and timing of therapy changes during the blistering cycles. Medication that does not urgently need to be changed, such as ‘gastric protection’ can be administered with the new blistering cycle, thus reducing unnecessary changes and additional work for care workers and the associated risk. It requires a certain amount of adaptation and flexibility on the part of the physician to avoid changing the therapy too often: I had the nurses tell me that and then therapy changes that are not urgent, I think about making them at that point. (NH3-PH). Or: ‘… should I prescribe it in 14 days, when I know that the next delivery (of the blistered medications) is coming … I always must consider that. And I mean, it’s just distracting. It just takes up resources.’ (NH1-PH).

Category 6: producing waste

From the perspective of the participants, waste production is seen by most as the only disadvantage of the new blistering system. The waste is generated from the blister bags, which are made of recycled material (plastic), but are thrown away after the medication has been administered and are not recycled yet: ‘The only disadvantage I can see is the rubbish. Because we produce a lot more rubbish now. Of course, with all the sachets. The cups I can rinse, and I can rinse the old blister boxes. The (new) blister bags are just a pile of waste’ (NH4-CNA4)

Even though some of the medication already blister-packed in the old system was disposed of, the fact that not only the bags but also the medication is thrown away when it is no longer needed, for example, in the event of a change in therapy within the blistering cycle period, is seen as a disadvantage of the new system.

Another example of waste production described that hospitalised residents cannot take the blister bags with them (it is not allowed by hospital policy) and must therefore be disposed of completely. ‘Yes, I don’t know how you can do it, but the point is, of course, that if I send a resident to the hospital, then of course, at the very least, up to 13 days of blistered medicines would be thrown away’ (NH1-PH1). A nurse manager put the statement into perspective: ‘Although, when they (the drugs) were prepared manually for a week, we also threw them away (NH3-NM).

Discussion

This qualitative study is one of the few to explore how NH staff experienced the implementation of an automated blister packaging system for solid, oral ingestible drugs. All four pilot NHs managed to implement the new system, and care workers, NH directors, nursing managers and physicians reported overall positive experiences. The automated medication blister packaging system was perceived by participants as having a positive impact on medication safety, although sources of error cannot be fully eliminated. Moreover, it created free spaces (for direct and indirect care) and reduced the workload of RN. The implementation generated efforts as workflows had to be adapted, and activities and thus roles and responsibilities were reorganised among different staff, which caused uncertainties in some nurses. The statement by one study participant ‘it pays off’ summarises that considering the staff’s efforts in implementing and adapting to the new system several expectations were met.

Improved medication safety was a key expectation towards the new system. Overall, the respondents reported that the automated blister packaging system led to greater medication safety by reducing errors in prescribing and blister packaging. This is in line with other research findings for inpatients, although not specifically for NH settings. For example, a systematic review highlighted the overall positive impact of automated in-hospital pharmacy dispensing regarding clinical (safety) and economic outcomes.⁷ As the Institute for Quality and Efficiency in Healthcare (IQWiG) in a rapid report concluded that there is insufficient evidence regarding benefits or harm of automated compared with non-automated drug blistering systems, our findings contribute to building a better knowledge base regarding the use of such technology in the NH context.

Our study also identified new detailed (positive) safety aspects of implementing an automated blister packaging system in NHs, which are, to the best of our knowledge, not yet highlighted in literature. RN emphasised the completeness of the therapy plan and the easy ordering of the drugs through the specific software. At the same time, physicians mentioned positively the software’s web application (app) for prescribing the therapy. The software also acts as a source of information for care workers, as they can check the therapy plans and how each drug looks (through a photo). For the interviewees, user-friendliness of the blister bags was also important, reflected in the fact, that not care-depending/autonomous residents can open the blister bag by themselves and take the medication on their own. Additionally, care workers felt much safer thanks to the new system, which they can trust. Especially CNA and NA mentioned that they can now rely on safe, machine-produced blister bags containing the right medication in the right dose for the right resident. The safety potential of technical solutions such as an automated blister packaging system thus can be a way to balance medication risk areas related to staff with lower educational levels,¹³ which might be especially important for settings, where few qualified personnel work, such as NHs.¹⁴ Considering the decreasing number of RN working in NHs, an automated blister packaging system might be an important strategy for improving and ensuring medication safety in these institutions.

The results of the study showed that there remain sources of error. For example, it is still possible that care workers administer the wrong blister bag to the wrong resident or/and at the wrong time, as not the entire medication process has been automated. Even new error sources can be introduced in implementing and inadequately adapting to the new process. Batson et al underscore our findings by suggesting that particular attention should be paid to the optimal implementation of automated blister packaging systems. Therapy changes during the blistering cycle can lead to mistakes, as nurses must cut open blister bags to remove discontinued medication, perhaps for seven or more days till the new blister bags arrive from the blister centre. A particular aspect of our findings relates to the need for nurses and physicians to communicate about the necessity and timing of ‘therapy changes’. On the one hand, some therapy changes can be avoided or better timed in this way, for example, reducing the risks of nurses manoeuvring blister bags or even positively influencing the problem of overprescribing. On the other hand, it might be perceived that physicians have to subordinate their clinical expertise in prescribing medications to a machine/system, which could have a negative impact on resident outcomes. Both aspects, however, need further exploration.

Acknowledging the limitations of technology, the need to adapt work processes which might again be error-prone, and that human error cannot be avoided, measures should be considered to ensure that the new system does not create a ‘false sense of security’. Managers, physicians and care workers should be able to learn from errors or near-misses that continue to occur even with or despite the new system.¹³ It is worth noting, that in the present study, risk management/structured patient safety activities such as analysing medication incidents or risky situations, or utilising a critical incident reporting and learning system, were not mentioned. The absence of risk management strategies in the present study is in line with the scientific literature, which highlights that NH settings still pay too little attention to establishing a safety culture and applying methods to increase it.¹⁴ While such reporting systems are widely used in hospitals, little is known about the NH setting.^{14 15} For instance, the study of Bengtsson et al, which inquired into errors linked to medication management in NHs, recommends applying the ‘Man, Technology, and Organization concept (MOT)’ for systematically analysing critical events or near misses.¹³ Risk management tools should therefore be applied in NHs in addition to the implementation of safe automated blister packaging systems. Another important finding of this study concerns RN’ uncertainties. Implementing an automated blister packaging system affects RN’ activities, roles and responsibilities directly, as they have a key role in medication management in NHs. Some feared that automated blistering could lead to a loss of knowledge regarding medication and the overview of residents’ therapy plans. As the participants mentioned, RN need to know the residents’ therapies well, as their observations and suggestions are important for physicians to adapt or change the therapy. Insecurities in this respect should not be undervalued, but at this point, it is not clear if the perceived loss of overview and knowledge is related to losing competencies, or to the change of familiar procedures and tasks in general or both. In its report on patient-specific blistering, the Institute for Quality and Efficiency in Healthcare points out that the effects of automated blistering on the professional competence of care workers are not clear and that there is a need for research in this area.⁶

Nurses’ expressed uncertainties could also be due to a lack of skills in dealing with change. Complexity theories suggest that change must be managed positively and that developing adaptive capabilities in staff should not be forgotten.¹⁶ Thus, gaining a better understanding of RN’ uncertainties concerning the implementation of technological solutions, such as automated blister packaging systems, needs further investigation. Moreover, RN’ uncertainties might be related to their changing tasks, roles and responsibilities in NHs. It is especially important that managers and clinical leaders create preconditions for their staff to feel confident to adapt and continuously improve, as implementing new technology and organisational change is transgressive and always interferes with work rules and processes, but also with people’s routines and (implicit) expectations of their role.¹⁷ The implementation of the automated blister packaging system can be seen as a chance and opportunity for management and policymakers to clarify such roles and responsibilities between care workers even at a regional level, and not just for a single structure.

Although the interviewees did not report any major time savings, they did experience a reduced workload and more free time for other nursing care activities. This is an important finding, which was not yet reported by other studies. RN described that there is now more time for other indirect care activities, which were sometimes left undone, such as updating care plans, following up on ward rounds and monitoring. This result can be seen as countering those positions where nurses felt deprived of tasks due to the machine blistering. The aspect of ‘taking the pressure off’ and thus reducing the burden and stress level that existed for RN on the day of manual blistering is a factor that should not be underestimated in connection with job satisfaction. Numerous studies emphasise the need to seek solutions or implement technologies that reduce workload so that employees stay on the job and do not leave.^{16 18} In addition, the positive realisation of the new blister packaging system takes some of the pressure off management to find and employ enough nurses. It is presumably also motivating to remain open to other technological innovations due to the positive experience of the people involved. One of the main reasons for introducing the new system in the NHs was to shift the workload related to drug blistering to a machine/robot, which was expected to save time for the now scarce nurse resource. The question of how much working time can be saved by automated blistering cannot be determined based on this study as it was not in its scope. Further research is needed to evaluate whether nurses perceived work relief due to the new system can be maintained over a longer period.

Our findings underscore that the implementation process of an automated blister packaging system in NHs is complex and time-consuming, as highlighted also by Hänninen et al.⁵ Despite NH staffs’ perceived positive impact on medication safety and the overall satisfaction with the new system, the following improvement areas were identified: creating interfaces/integrations of SiCuro software with the nursing documentation software (ie, SENSO^® 7); clarifying roles and duties of different care workers, as well as standardising medication management workflows within the NH; and reducing the risk surrounding ‘therapy changes’.

Strengths and limitations

This qualitative study makes an important contribution to an under-researched area of how implementing new technologies such as an automated blister packaging system in NHs is experienced by staff. The strength of this study is that data were collected by seeking perspectives from various points of view, for example, from NH directors to NA, through exploring their perceptions and experiences. The extent to which the four volunteering pilot homes are representative of all NHs in the region due to their high willingness to innovate and motivation to try out this new technology remains unclear, which is why this factor must be considered for the successful expansion of an automated blister packaging system to other homes. However, the fact that the new automated blister packaging system was implemented in the four NHs with mainly positive experiences despite the differences in structure (eg, size, number of residents, staffing levels and mix) and approach to implementation (eg, type of process adaptation, degree of standardisation within each NH) can be seen as encouraging for the transferability of the results to other NHs. Furthermore, it is acknowledged that quantitative enquiry is needed to confirm results concerning improved safety and efficiency aspects, as the findings of this study explored NH staff’s perceptions and experiences only.

Conclusion

With this qualitative study, we aimed to evaluate the NHs staff’s experiences with the implementation of a centralised automated medication blister packaging system for solid, oral ingestible drugs in four NHs in South Tyrol (Northern Italy). The experiences were summarised along six main categories, namely impacting safety, creating free spaces, serving residents, meeting expectations, generating efforts and producing waste. According to NH staff, the automated blister packaging system has led to improved medication safety and reduced the workload of RN, but it is not an ‘all-around carefree package’. Implementing the new system provides also opportunities to clarify roles and duties among the different care workers . However, continuous quality improvement and risk management measures are recommended to accompany implementation and its sustainment. Further research is needed to gain a better understanding of the impact of an automated blister packaging system on RN’ competencies in NHs regarding medication management and residents’ therapies and safety.

Data availability statement

Data are available upon reasonable request.