Older persons’ experiences regarding evaluation of their medication treatment—An interview study in Sweden

Abstract Background Older persons with polypharmacy are at increased risk of harm from medications, and this issue is a global patient safety challenge. Harm may arise at all stages of medication use and may cause hospital admission, additional resource utilization and lower patient satisfaction. Older persons’ participation in their own care may increase patient safety. Their views on the evaluation of their medication treatment, and their own involvement in it, are crucial yet poorly understood. Objectives To identify opportunities to make the medication use process safer, we explored and described older persons’ experiences of evaluation of their medication treatment. Design Semi‐structured interviews were performed with 20 community‐dwelling older persons (age 75‐91 years) in Sweden. Data were analysed using inductive qualitative content analysis. Results These older persons reported having a responsibility to engage in their medication evaluations, although some felt unable to do so or considered themselves unconcerned. Evaluation, in their experience, was facilitated by continuity of care and an invitation to participate. However, some older persons experienced not receiving a comprehensive medication evaluation. Discussion and conclusion Older persons want to be actively involved in their medication evaluations, and this may represent an underutilized resource in the pursuit of patient safety. Their trust in physicians to undertake evaluations on a regular basis, although that does not necessarily occur, may cause harm. Patient safety could benefit from a co‐production approach to medication evaluations, with health‐care professionals explicitly sharing information with older persons and agreeing on responsibilities related to on‐going medication treatment.


| BACKG ROU N D
The World Health Organization (WHO) identifies 'Medication without harm' as the third global patient safety challenge. 1 Harm from medications, often called adverse drug events (ADEs), occurs at all stages in the medication use process and cause hospital admission, additional resource utilization and lower patient satisfaction. 2 In a review of studies concerning preventable ADEs in ambulatory care among community-dwelling persons, the median prevalence rate was 16.5%, with a higher rate among older persons. 3 This finding highlights the need for efforts to reduce ADEs, especially in older persons. Improving the medication use process, including prescribing, preparing, dispensing, administering, monitoring and evaluating, may reduce or prevent ADEs 4 For older persons, inappropriate polypharmacy is a leading cause of ADEs. 5 Studies with different interventions to reduce inappropriate polypharmacy, such as medication review or assessment of medications, have had difficulties in demonstrating clinically significant improvements. 6,7 ADEs are common but often preventable in older persons in ambulatory settings.
These events often originate in the prescribing or monitoring stage. 8 Consequently, interventions aimed at these stages seem most beneficial in the prevention of ADEs. Traditionally, monitoring involves performing a physical examination and reviewing the patient's history and laboratory tests. An evaluation based on the results of this monitoring informs decisions about future treatment. 4 Here we define 'evaluation' as an assessment of performance against an established set of goals or objectives at a point in time. 4,9,10 Evaluation is continuously relevant, not only at fixed intervals, because medical conditions, especially for older persons, can change rapidly.
Regulations by The National Board of Health and Welfare in Sweden direct prescribers to document plans for when and how a treatment will be evaluated. 10 Older persons themselves, their relatives and health-care professionals play significant roles in the medication use process. 11 Persons participating in their own care may help reduce the risk of errors and increase patient safety, likely because they are involved, and vigilant persons can observe and communicate problems they experience before these problems result in adverse events. 12,13 While many older persons want to participate in their own care, 14 not all have the desire or ability to do so. Older persons' views on the evaluation of their medication treatment, and their own involvement in it, are crucial yet poorly understood. To generate opportunities for improvements to make the medication use process safer, it is important to know how these individuals currently experience the evaluation. Therefore, the present study explored and described older persons' experiences of evaluation of their medication treatment.

| Recruitment and data collection
For this interview study, we recruited older community-dwelling persons, with at least one chronic disease treated with medication(s) on a regular basis. We used the 'older age' definition, age 75 years or older, from the national regulation on medication use. 10 To access a variety of experiences, a purposeful sampling was used and included persons with different numbers of diagnoses and medications, sex and age. Persons who managed their medications by themselves and those with support from a close relative were included. Persons who were not able to speak or understand Swedish and those diagnosed with dementia were excluded. Health-care staff at five primary care centres in southern Sweden recruited the participants. Centres were selected to include large and small, publicly and privately run, centres. Staff verbally informed potential interviewees about the study and asked about participation. If the person wanted to participate, one author (MH) was notified and contacted the person within one week to give further information about the study. If the person agreed to participate, an interview was scheduled. All invited persons consented to participate.
One author (MH) performed all the semi-structured individual interviews, guided by an interview guide that focused on the older person's experiences of evaluations of their medication treatment.
The interview guide was developed based on knowledge on patient safety among the authors and guided by the medication use model 4 to cover different aspects of evaluation. It was pilot tested with three participants which yielded minor clarifying adjustments. The content of the pilot interviews did not differ from subsequent interviews, so these pilot interviews were included in the analysis. Each interview started with the invitation 'Describe your latest appointment with a physician, in which you talked about your medication they were asked to choose the time and place for the interview. 15,16 Most participants (n = 16) chose to be interviewed at home, and the others (n = 4) chose their primary care centre. The interviews were audio-recorded and lasted between 28 and 65 (mean 47) minutes.
After completing 20 interviews, no new information emerged, suggesting that sufficient data were collected to describe the phenomena. Therefore, no further participants were included.
Demographic characteristics were collected from the participants and from their medical record, with their consent (Table 1).

| Data analysis
Two of the authors (MH and LJ) performed qualitative content analysis according to Elo and Kyngäs. 17 As there is little previous knowledge in this area, the analysis followed an inductive approach. 17,18 First, the interviews were transcribed verbatim and read several times to become familiar with the data. Open coding involved writing headings in the margins and putting all headings together in a coding sheet. Data headings with similar content were compared and grouped together to generate subcategories. By abstraction, similar subcategories were formed into generic categories that were divided into main categories and reflected the content of the interviews. Examples of the analysis process are presented in Tables   2 and 3. Data were analysed and discussed among all authors until a consensus was reached. After data were analysed, the interviews were read through one more time to validate the categories that emerged. The preliminary result was then discussed with the other authors.

| RE SULTS
The analysis of older persons' experiences formed two main categories that described their own role in the evaluation and their views of evaluation received. Each main category included three generic categories.
The first main category, participants' own role in the evaluation To help make sense of these findings, we elaborate on our analysis of the main and generic categories and present illustrative interview quotes.
TA B L E 2 Example of the analytical process-One generic category in the main category 'Own role in the evaluation'

| Feeling unable
The older persons described feeling unable to participate in the evaluation of their own medication treatment in several ways, including having trouble remembering verbal information about treatment goals, test results or how the treatment was supposed to be evaluated, especially if they received too much verbal information at one consultation. These memory challenges made them uncertain of whether they had received that information and what actions they were supposed to take.
Yes, some things one remembers ---but it can be like stuffing too much information in, so to say. When you sit and go through a list like this, you know, and you concentrate ---there may be something that gets lost, you know. (P2) The older persons sometimes found it difficult to interpret the medication effectiveness and potential side-effects by themselves.
Knowing whether the current treatment was the right treatment for them was difficult because they were unsure of how they would feel without the medications.

| Being unconcerned
Being unconcerned about the evaluation reflected older persons' experiences that taking medications was not stressful or worrying, but essential. Another reason for the lack of concern was related to age. Some persons accepted adverse reactions because of their age or because they thought that medications at their age were intended more to relieve symptoms than to cure the underlying condition.
Additionally, being unconcerned about the evaluation was related to the older person's trust in the physician to have the knowledge needed to perform the task properly. As long as they were feeling fine and did not experience side-effects or worsening of symptoms, they did not question the way the evaluation was performed and saw no need for monitoring visits. The older persons expressed that 'no news is good news' and presumed everything was fine if they did not hear anything. They expected that treatment was under control if they did not receive any information of concern about test results or directions to change the treatment.
And then I go and take blood tests. And I have never heard anything so it must be fine then. well in other aspects. Although the intended duration of treatment in many instances was not known, it was not described as a cause for concern. A medication was sometimes continued without knowing why it was prescribed and without questioning its use. They interpreted a renewed prescription as a signal that the medication was evaluated and considered appropriate for another period of time, even if it had initially been intended for a finite period.

| Obtaining continuity
Obtaining continuity of care with a structured evaluation of their condition and treatment on a regular basis, for example via a yearly visit to their general practitioner (GP), was appreciated and facilitated the medication evaluation. Typically, older persons received written invitations to these visits, and the invitations were expected.
They emphasized that seeing the same physician, who knew them and their medical history, resulted in a sense of continuity and safety. Yes

| Being invited to participate
When a physician or nurse invited them to discuss their medication treatment, the older persons felt safe and involved. Having a dedicated person to contact at a primary care or specialist care centre made it easier to ask questions or request an appointment for an evaluation.
Being involved in the evaluation of medications was experienced when the physician shared results from blood tests, blood pressure readings, diagnostic imaging or other tests and gave advice on con-

| Lacking a comprehensive evaluation
Older persons sometimes expressed concerns about the lack of a comprehensive evaluation of all their medications taken together.
Those who received prescriptions from several physicians missed having someone explicitly evaluate whether the mix of all medications was appropriate, and whether the purposes of the prescribed medications were achieved. Sometimes, they felt that the effects of some medications opposed those of other medications.
Well, but it's not easy to know. They [the physicians] say, some say 'Yes, I have to have a pill to calm me down'. Yes, yes, then they will get one. 'I must have one that perks me up' ---Yes, then they get one for that. And then, they don't know which is which ---or whether they work or not…' Older persons who had contact with other physicians in addition their GP described a lack of cooperation between them. For example, medication initiated by a medical specialist was later evaluated by the GP on referral, but the GP was not familiar with the treatment, and it was unclear to the older person who was responsible for the evaluation. Knowing who to ask for further evaluation sometimes seemed difficult because physicians rotate when they are under training or medical locums.
The older persons questioned the practice when physicians renewed their prescriptions without an appointment, because the physician may lack up-to-date information about medication effectiveness, for example a recent blood pressure reading. Some participants stated that visits lacked sufficient time, and it was difficult to secure time to discuss the treatment with the physician, especially its purpose and potential side-effects.
No, no, nothing, but she said 'You should take this pill' and I had no idea what it was. ---She could have sat down [bedside] and said that this one is for this and that and so on, but no… ---It's not… they do not have time for that. (P14) A lack of information about how, when and by whom treatment was supposed to be evaluated was expressed. Even if health-care professionals sometimes invited the older persons to monitor their treatment themselves, the results were not always requested. When receiving results, for example from laboratory tests or diagnostic imaging, often no findings were specified; they were just told 'everything is fine', although the older persons wanted to receive more specific information.
Yes of course. If you go and take blood tests, then you ought, at least, to get some kind of result, one would think. ---Because the doctor, he gets the test results, but I do not. (P18) The frequency of monitoring and evaluation varied over time and between different treatments. When medications were initiated or changed, frequent evaluation with visits or phone calls was received, but medications were not evaluated as frequently over time. For some older persons, years passed between assessments, which they questioned. They reasoned that this lack of evaluation was related to their age, that it may not be needed or that treatment monitoring was omitted for older persons simply because they were older.

| Reflections on the findings
The older persons experienced their medication treatment as safe, but they wished that a designated health-care professional took the overarching responsibility for their medication evaluations.
Their trust in the physicians and their desire to be involved in their medication treatment are two important findings that must be addressed. From a safety perspective, it is important to identify and strengthen actions that make patients feel safe but also 'remain sensitive to possibilities of failure', so failures are detected and addressed before they become ADEs. 19 The older persons revealed a deep trust in physicians and assumed that these physicians took responsibility for the evaluation of their medications. This trust in physicians' medical knowledge on whether to proceed or discontinue a medication mirrored findings about older persons' attitudes towards de-prescribing. 20  Older persons' deep trust in the physicians, combined with a lack of concern about risks with their medication treatment, may make them less prone to notice and report potential safety hazards. To prevent inappropriate polypharmacy and ADEs in older persons, these findings highlight the importance to take a systematic approach to assess and improve medication treatment, which requires cooperation between health-care professionals and patients. 23 This study shows that there are older persons who want to be involved in the evaluation of their medications, just as in care in general. 14 However, it also revealed several challenges to patient involvement, including insufficient time at appointments to discuss on-going treatment, difficulties in understanding or remembering information, and lack of written information. This corresponds well with factors found to make older persons feel insecure about their medications. 24 In Australian nursing homes, residents who lacked understanding of the purpose of their treatment, or risk of potential ADEs, tended to develop apathy towards the risks related to polypharmacy. 25 In a recent survey, people in Sweden reported having less time for consultation with a physician and receiving less information about their treatment compared with people in other countries. 26 Sufficient consultation time, written information and good access to health-care professionals to ask questions are important considerations to ensure that older persons understand how their medication treatment is supposed to be evaluated, which is important for their ability to participate and enhance safety.
Persons who saw themselves as responsible for their medication evaluation monitored their treatment at home and proactively scheduled follow-up appointments with their physician. To be involved and self-reliant in the medication use process can, as shown in studies of self-administration, make people feel safe. 27,28 Giving older persons support to be more active in their health care, for example during transitions such as discharge from hospital and discussing symptoms and adverse reactions to be alert to, may reduce the risk of undesired consequences such as readmission. 29 An active partnership between physicians prescribing medications and the person taking them (or not) exemplifies the co-production of health and health care. 30

| Methodological considerations
The major strength of this study is that it accessed the experience of medication treatment and evaluation in older persons, in depth, using semi-structured individual interviews. To promote the trustworthiness of the study, we considered the credibility, dependability, confirmability and transferability of the data collection and analysis. 36 One author (MH) performed all interviews, which promoted consistency across interviews. We used purposive sampling to access a range of relevant experience among the participants and achieve transferability. Primary care centres identified suitable participants on the authors' behalf. We note that 25% (n = 5/20) of the participants had worked in health care, as nurse assistants, nurses or pharmacists. This experience may have provided these persons with a greater understanding of the medication use process than among older persons in general. This experience also likely provided these participants a good ability to discuss medication evaluation.
Selection of sampling size is important to ensure credibility. 37 A preliminary analysis of 20 interviews grouped the data and created concepts, and therefore, we accepted the number of collected interviews. To achieve dependability in selecting relevant meaning units and form categories that covered the data, two authors (MH and LJ) analysed the material together. They then discussed preliminary results together with the other authors, combining perspectives from pharmacy, nursing, gerontology, medicine, patient safety, improvement research and qualitative methods to reach consensus.
The inclusion and exclusion criteria and the purposive sampling method created dependability in the recruitment process. 37 After the categories emerged, all interviews were read through one more time to ensure the confirmability of the participants' experiences of evaluation.

| CON CLUS ION
Older persons' experiences of the evaluation of their medications reveal several opportunities to improve medication treatment safety. Many older persons can and want to be actively involved in their medication evaluation, and this desire may represent an underutilized resource in the medication use process. However, their trust in physicians to undertake evaluations on a regular basis, although that does not necessarily occur, may cause harm.
From a patient safety perspective, it appears that older persons will benefit from a co-production approach to medication evaluations, with health-care professionals explicitly sharing information with them and agreeing on responsibilities related to on-going medication treatment.

ACK N OWLED G EM ENT
The authors thank the older persons who participated in the interviews and health-care staff in primary care who helped out with the recruitment.