Article Text

Original research
Health literacy in medication communication during hospital discharge: a qualitative study at an internal medicines ward in Norway
  1. Kajsa Rebecka Bengtsson1,
  2. Stine Eidhammer Rognan2,
  3. Sofia Kälvemark Sporrong3,
  4. Helene Berg Lie1,
  5. Yvonne Andersson4,
  6. Morten Mowe5,6,
  7. Liv Mathiesen1
  1. 1Department of Pharmacy, University of Oslo, Oslo, Norway
  2. 2Department of Pharmaceutical Services, Oslo Hospital Pharmacy, Hospital Pharmacies Enterprise, South Eastern Norway, Oslo, Norway
  3. 3Pharmacy Practice and Policy, Institution of Pharmacy, Uppsala, Sweden
  4. 4Hospital Pharmacies Enterprise, South Eastern Norway, Oslo, Norway
  5. 5General Internal Medicine Ward, the Medical Clinic, Oslo University Hospital, Oslo, Norway
  6. 6Faculty of Medicine, University of Oslo, Oslo, Norway
  1. Correspondence to Kajsa Rebecka Bengtsson; kajsarebeckabengtsson{at}


Objective When discharged from hospital patients are often assumed to have sufficient health literacy (HL) to participate in their medical treatment and manage medical self-care after discharge. However, limited HL is a widespread concern and patient participation during discharge is lacking. In this study, we explore how HL influences medication communication during hospital discharge.

Design A qualitative case study, comprising unstructured observations of patient–healthcare personnel (HCP) encounters followed by semistructured interviews. Data were analysed using content analysis.

Setting An internal medicines ward at a university hospital in Norway.

Participant Fifteen patients aged 40–89 years were included close to the day of discharge.

Results The following themes describing dimensions of HL emerged: (1) access, (2) understand, (3) appraise and (4) apply. Most patients sought access to medication information from HCP, while some felt dependent on HCP to provide it. However, their abilities to understand, evaluate and make informed decisions were challenged, partly because HCPs’ ability to adapt their communication to the patient’s knowledgebase varied.

Conclusion The results give a broader understanding of how HL influences medication communication during hospital discharge. To consider central dimensions of HL is important to achieve optimal medication communication, as the communication only can be exercised within the frames of the patient’s HL. The findings in this study support that HL should be described as a shared responsibility between the patients and HCP. Attention should be focused to the HCP’s responsibility to adapt the communication to the patient’s knowledgebase.

  • quality in health care
  • medical education & training
  • public health

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Duo to the sensitive nature of data in this study, patients and HCP were assured raw data would remain confidential and would not be shared.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

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.

Strengths and limitations of this study

  • Combination of observations and patient interviews is a powerful approach that describes what happened from different point of views, which is a strength to this study.

  • The sample size is limited; however, the rich interview data and observations provide a high information power.

  • Participation was voluntary and a bias towards empowered and confident patients cannot be excluded.

  • This study was performed at one internal medicines ward located at one hospital and the transferability can be questioned.


Medication communication with the patient during hospital discharge is often insufficient, potentially causing adverse drug reactions, medication discrepancies or hospital readmission.1 2

According to the World Health Organization (WHO) high health literacy (HL) empowers the patient to engage in decision-making about their health, and WHO recognises HL as a ‘critical determination of health’.3 However, limited HL is a widespread concern, associated with a poor ability to comply to a medical treatment, a decreased use of preventive healthcare services and higher hospitalisation rates.4–6 HL is an evolving concept commonly described as the skills determining the patient’s ability to gain access to, understand and use information in ways that promote and maintain health.7 Newer definitions describes HL as a dual sided concept, where there is a shared responsibility between the patient and the healthcare personnel (HCP).8 9

The focus of healthcare systems is shifting from the conventional way of practising medicine towards an ambition to involve the patient in the decision-making.10–12 In this shift, the healthcare systems are assuming that the patient has sufficient skills to comprehend and use health information, that is, adequate HL, and that home-dwelling patients self-manage their medications after hospital discharge.11 13 14 Despite it being well known that clear and understandable communication empowers the patient, HCP frequently presents health information at an HL level higher than the patient can comprehend.13 15 16 Previous studies have identified the patient’s engagement during hospital discharge to be lacking and home-dwelling patients are frequently reporting difficulties in medical self-management and in the understanding of health information provided from the hospital.17–19

The patient is the only constant through the ‘patient journey’ and should be recognised as the key actor in their healthcare.20 While extensive research has been made to improve hospital discharge, it is essential to understand the patient’s perspective in addressing health information during and after discharge.19 By observing the hospital discharge and follow-up with patient interviews, the medication communication can be captured and described from different point of views.21 In this substudy, which is a part of a broader research project (see online supplemental file 1), we explore how HL influences medication communication during hospital discharge.


Patient and public involvement

A representative from the User’s Board of the Hospital Pharmacies Enterprise provided input to the study protocol (see online supplemental file 1) and ensured that the information sheet to be handed out and explained to the patients provided a good summary of what the participants needed to know before signing the consent form. The user representative has more than 5 years of experience from the User’s Board and has a master’s degree in welfare management.

The Norwegian hospital setting

The patient’s right to sufficient health and medication information is anchored in Norwegian legislations. This includes the right to participate in the organisation of healthcare services, where the level of participation and information should be adapted to the patient.22

Hospitalised patients receive their medicines from the hospital during their stay, but do not get them dispensed at discharge. Home-dwelling patients who manage their medicines themselves will normally collect their prescriptions at a pharmacy of their own choice.


This substudy was conducted using a qualitative research design, comprising unstructured observations followed by semistructured interviews.

Sampling strategy and setting

The study setting was at an internal medicines ward at a university hospital in Norway. The observers (two pharmacy students and a clinical pharmacist, authors KRB, HBL, SER) enrolled hospitalised patients into the study from September to December 2019, close to the day of the patients planned discharge. Thereafter, the patients were observed during medication-relevant encounters with HCP, through to hospital discharge. After discharge, the patients were interviewed in their home, at a temporary sheltered unit, a café or by telephone.

Patients were included through purposive sampling. To ensure variation in demographic characteristics, eligible patients were selected, based on sociodemographics (eg, gender, age, education, ethnicity), diagnoses and assumed length of hospital stay. Eligible patients should be over 18 years old, home dwelling and expected to be discharged to their homes or a short-time nursing home department. Preterminal or cognitively impaired patients were not eligible.

Data collection

A pilot study was performed by KRB, HBL, SER, LM to ensure synchronised observations and to develop an observational form (see online supplemental file 2). In Norway, HCP at hospitals normally wears white uniforms. The observers disclosed their HCP background, but dressed to appear as ‘the girl from the university’, wearing a yellow t-shirt with the word ‘observer’ across the front, rather than HCP.23

Data were collected Monday to Friday from 08:00 to 15:30. In addition, observations were conducted over 1 weekend to gain a wider perspective of the patient’s experience. Communication involving medications were documented, along with descriptive data like environment, behaviours and the affect on patient activation. Patient’s demographics, medical treatment, medical history and discharge summary were collected from the medical records. All patients were mainly observed by one of the observers, in order to maintain continuity for both parties. A second observer stepped in when necessary, for example, during lunchbreaks. The observations were audio recorded if the patient stayed in a single room and both the patient and HCP gave their consent.

The interviews were conducted by KRB or HBL within 2 weeks after discharge and were audio recorded if the patient gave their consent. Interviews were performed using an interview guide (see online supplemental file 3), containing suggestions of open-ended questions and personalised to each patient based on data from the observations.

By continuously comparing eligible patients with data from previously enrolled patients sufficient information power was strived for.24 Saturation was appraised to be reached after 15 observed patients and 10 patient-interviews. Of patients approached, one declined to participate. The patients from the pilot study (n=3) were not interviewed, one patient declined to be interviewed and one patient was not reachable after discharge.


The research teams for the analysis had different backgrounds (education and experience in the hospital setting), which provided different perspectives. For the analysis presented in this article, data material from all observations and interviews was analysed. The initial step of the analysis was to inductively create codes using transcripts from the pilot study, first individually and later in multiple consensus sessions (KRB, HBL, SER, SKS, YA, LM). This resulted in a codebook used to deductively code the remaining transcripts (KRB, HBL, SER). Text from code groups involving medication communication and qualities reflecting HL was then condensed into units of meaning. Code groups were combined cross case and the identified units of meaning were clustered into themes, representing four dimensions of HL, inspired by Sørensen’s model of HL, table 1.9

Table 1

Four dimensions of health literacy9

The content in each theme was reduced into a condensate and modified into descriptions and interpretations in form of an analytic text and quotations. To maintain sociocultural context and ensure interpretative validity, translation into English was done after fulfilment of the analysis.

The patients are presented with age. The quotes are from observations if not specified with ‘int’.


Fifteen patients were included in the observational study, of which 10 were interviewed after discharge. Patient demographics are presented in table 2.

Table 2

Demographics for patients that participated in the study

The median length of observations was 2 days (range 1–22 days) and the interviews varied between 33 and 87 min (median 55 min). The collected data material consisted of 295 513 words. The thematic analysis comprises four themes; access, understand, appraise and apply.


All patients in this study received medication information during hospitalisation, typically when HCP handed out medicines, during ward rounds and in a written discharge summary. Nevertheless, the level of details often seemed to depend on what the patient had requested.

It appeared as most patients felt responsible to obtain sufficient medication information, by asking HCP questions. Some patients were observed to have written down their questions and thereby prepared themselves for the physician’s visit. However, in the interviews, some patients expressed that they wanted specific medication information during hospitalisation but felt dependent on HCP to provide it.

The HCP explained what medicines I received and why I should take them, but very little about side-effects. It would have been nice to get a heads-up about what to expect.

60, int

In the interviews most patients informed that they had been seeking medication information through online searches, by reading patient information leaflets or by asking for medication information at the pharmacy.

I checked a lot of details with the pharmacy staff when I picked up my medicines after discharge. There is one question I always ask; is there something I should have asked you, that I haven’t? It makes them think, because they are the ones with the knowledge.

58, int

Even patients who received assistance from next of kin or home care nurse expressed a need to remain some sense of control. However, it appeared as though a few elderly patients, not responsible for handling their medications after discharge, did not feel the need for medication information at the hospital.

I’m sure I received good medicines information, I just didn’t bother listening to it.

71, int

HCP was occasionally observed to hand out medicines without sharing information. In these situations, the medical treatment would be unknown to the patient if the patient did not request information.

I got the medicines required, at specific times, so I just took them and thought that’s the way it should be. There were no questions. I don’t have any clarity of what sort of medicine it was, but it was enough information for me.

89, int


The patients showed different ways of understanding medication information, as some patients appeared simply to acknowledge it, while others strived to derive meaning from it.

I got a prescription for sleeping pills that I haven’t used. I sleep better now, after I started with my heart medicine. It has slowed down my pulse, which I guess is what kept me up at night.

53, int

It was commonly found that the patients reassured that they understood the medicines information at the hospital by responding with questions or asking HCP to clarify or repeat information.

So, it’s not sure my blood sugar will continue to be high when I finish this cortisone-treatment? I understand diabetes is a common side-effect.


However, some patients were responding with a nod or humming in what seemed to be agreement when receiving medication information, making their level of understanding unclear. During interviews, examples of patients that had struggled to understand the information were revealed.

She used a physician’s-language that was hard to understand, so I was thinking: what is she actually saying now?

67, int

The patients often had difficulties in understanding their written discharge summary. Despite this, the written medication information appeared to be an information source for the majority of patients.

Like this, it says butenamid in my discharge summary, that is the same as Burinex. But it says that butenamid is for heart failure and I know Burinex is a diuretic… is it really the same then?

70, int


The patients appraised their own ability to evaluate medication information differently. They were commonly found to evaluate their medical treatment, both during hospitalisation and after discharge.

She [the physician] said I should take this medicine at night, because my blood pressure was peaking in the morning. But I’ve also read a report about the medicine saying it’s 50% more effective if you take it at night.

58, int

One patient evaluated HCPs’ information during hospitalisation to ensure correct medical treatment and appeared to trust herself more than the HCP.

The nurse told me to take the tablet a half hour after breakfast. But I thought that was wrong, that’s not what I’ve read. I googled it just to make sure I remembered correctly, and I did. You should take it on an empty stomach in the morning, 4–8 hours after food.

53, int

Even though the majority of the patients appeared to evaluate the medication information, the level of source criticism was found to be diverse. There appeared to be a prevalent use of online search engines, while only some claimed to use well known and trusted websites.

I don’t remember what website I used… I just googled the name of the medicine and read whatever came up.

60, int

On the other hand, it appeared as some patients found their own ability to evaluate medication information to be insufficient, implying that it is the physician’s responsibility.

I think… to be able to discuss anything about medicines, you have to be a doctor. I can’t decide anything. I just have to trust the doctor when he says that the medicine is good for me.

48, int


Some patients appeared to strive for control of their medical treatment, while others seemed to derive security from a ‘HCP know best’ attitude.

Patients who appeared to seek control often expressed their medical needs and challenged HCP regarding medical issues.

I needed medicine for my enlarged prostatic gland, and I alerted the HCP about difficulties to urinate during my whole hospitalisation. I really had to push for them to act on that.

60, int

However, situations were frequently observed in which the patients did not have the opportunity to be in control of their medical treatment, as they got informed about changes in their medical treatment after the decision had been made. Some patients tried to regain the control by not accepting the medical changes presented to them.

They said I should start on blood thinning medicine, but I said no. The doctor became a bit like, huh?! They did not expect to hear that. But then I said I would do it, I have to do it, right?

53, int

A few patients who felt uncertain about their medical treatment after discharge took responsibility by contacting the hospital to get an extra assurance.

I called the hospital after discharge. It was only pills for the morning and evening in my medicine-dispenser, not for mid-day as I got in the hospital. It didn’t say anything about this in my discharge summary either. I was wondering if I had been taken of it or if it was wrong that I didn’t get it anymore.

67, int

Some patients had made changes to their medical treatment, without consulting HCP. One patient did not comply with the dosage of his inhalator from the discharge summary, as he stated that the information in the discharge summary was incorrect. Another patient’s cause of admission was because of his intentional discontinuation of prescribed medication:

I used blood thinning medicine before, but I got side-effects and it didn’t seem to have any effect on my actual problem. I stopped using it on my own initiative, which I guess was a bit stupid of me.

60, int

Some patients appeared to take a more passive position, not striving for control in their medical treatment, assuming that HCP was in charge of all decisions. An example of this is one patient who complied with her medical treatment even though she knew it included drug–drug interactions (DDIs) and did not suit her needs.

At the pharmacy they told me that the new medicine did not fit with another of my medicines. This was a Friday, I took the pills during the weekend, but when I mentioned it at my scheduled appointment with the GP on the following Monday, he called the hospital immediately.

I asked if I could take it [nitroglycerine-spray] before I went out, but the doctor said that I only should take it when I feel chest pain. Then I have to take it in public, which I think is dreadful, but I follow doctors’ orders.

83, int

However, even for the more passive patients, it still seemed important to engage, for example, to make sure that the practical aspects of medication management were taken care of.

Everything is new to me, it’s a bit of a fuzz. If you could fill the pill-dispenser until Thursday, the home nurse services will take care of it after that. Will you notify the pharmacy?



This study aimed to explore how HL influences medication communication during hospital discharge. Previous studies often focus on the individual patient’s HL; how to measure it and correlate it to the patient’s health outcomes and skills in health communication or health behaviour.5 6 25–32 The findings from our study describes HL as a shared responsibility between the patient and HCP and supports HL to be a dual sided concept, as high-quality medication communication is essential for the patient to play an active role in their medical treatment.8 9

The patients in our study either took on an active or passive role in their medical treatment.15 33 The active patients wanted to engage in their medical treatment and appeared motivated to seek information, while passive patient appeared more dependent on the HCP making decision for them. Low and high HL levels have been connected to patient’s characteristics like being passive or inactive in their approach to healthcare versus being active and feeling in control about their healthcare.34

HCP did not always adapt their verbal and written communication to the patient’s knowledgebase, creating a gap in the information provided and perceived. Previous studies show that HCP often overestimates the patient’s HL and uses a medical jargon incomprehensible to the patient.31 This might indicate that the patient is expected to learn areas of expertise that are defined by the healthcare system and that HL is a concept that should not belong solely to the patient.8

Some patients in our study were seeking medication information online, which may indicate dissatisfaction with patient–HCP medication communication.35 In addition, many patients appeared to lack awareness of relevant and reliable websites, as search engines were often the first port of call. Previous studies problematise online health information accessible for patients, as patients may not understand or that an information overload may cause the patients to feel less empowered.27 35 Generally, in our study patients still valued the advice from HCP above online health information. HCP could give guidance to the patients in using quality online information sources.35

The passive role of some patients in this study may be a consequence of not understanding or a fear of uncovering the knowledge gap. To accept information without question or without seeking to ensure that it meets ones needs is associated with inadequate HL.34 These qualities can also be correlated with the patients external health locos of control (EHLOC), for example, relying on that HCP know best rather than one’s own capability of being in control of one’s health.36 An example of this is described as one patient complied to her medical treatment even though it did not suit her needs and she knew it included DDIs. However, EHLOC fosters the patients’ acknowledgement of HCP’s medical decisions, which can be beneficial to health and is therefore not equal to inadequate HL.36 Previous studies have concluded that patients that choose to depend on HCP to make health decisions on their behalf often are capable of making informed health decisions themselves.14 The question is if the patients in our study were silent by choice or silenced due to lack of knowledge.

Patients who believe that their health is directly related to their own actions, internal health locos of control (IHLOC), are assumed to more likely engage in a healthy behaviour.36 The results from this study support that this assumption is true. However, one of the patients who could be described as active, and possessing IHLOC, had discontinued life-necessary medicines on his own initiative, resulting in the current hospitalisation. This patient also struggled to understand the medication information provided from HCP. This exemplifies that IHLOC and an active patient role do not always correlate with adequate HL or healthy outcomes. Adequate HL requires more than to actively seek medication information; it is a key ability to identify when to act autonomously and when to ask HCP for guidance.36 37 One can argue that on this occasion the healthcare system failed to communicate understandable health information to the patient, and to assume that this patient’s HL is inadequate would seem unfair.

In this study, patients were frequently observed not to have an opportunity to be in control of their medical treatment, as decisions were made in their absence. Even though a few patients argued against a decision that had been made by HCP on their behalf, the general approach appeared to be acceptance, that is, engaging in a passive patient role. Several studies suggest that a lower ambition to participate in decision-making regarding health is related to inadequate HL.25 28 38 39 However, similar to our observations, it is argued that patient participation rather emerges from HCP giving patients an opportunity to participate.40 41 Previous studies suggest that patients adjust to the HCP’s level of engagement, as low HCP engagement seems to trigger an active patient role and vice versa.15 However, studies also indicate that HCP more often take the initiative to invite passive patients to participate in medical decisions.41 This implies that the level of engagement among the patients in our study may have been affected by the HCPs behaviour designated to increase patient activation, and not solely by their own will.40

The combination of observations and patient interviews is a powerful approach that captures and describes what happened from different point of views, which is a strength to this study.21 The patient experience does not always equal the objective story. Examples of this were one patient who described a 30-min long discharge conversation that was observed to last for only 10 min.

Participation in the study was voluntary and a bias towards empowered and confident patients can therefore not be excluded. However, the vast majority of patients asked to participate gave their consent and the recruitment of a heterogeneous sample of participants is therefore considered successful. Although the sample size in this study is limited, the rich interview data and observations provide a high information power.24

Interviews with patients and HCP (unpublished focus groups interviews with HCP) found that the observer did not affect the medication communication, at least not in the long run. The long observational time (4 months) at the setting, a hospital ward used to having, for example, students as observers, probably reduced potential observer effects.42 Furthermore, the observations were mainly conducted by one observer, who also did the interview, which helped build a relationship between the patient and the researcher. By identifying as the girl from the university rather than HCP, this relationship may have affected the patients to speak more freely in their interviews about their hospital experiences.23

Although the research team consisted of persons with different backgrounds, all but one were women and all had a Northern-European background. To limit the affect of potential preunderstanding bias, the researchers discussed their sociocultar position and value system during the research process.

This study was performed at one internal medicines ward located at one hospital and the transferability can be questioned.


The results give a broader understanding of how HL influences medication communication during hospital discharge. To consider central dimensions of HL is important to achieve optimal medication communication, as the communication only can be exercised within the frames of the patient’s HL. The findings in this study support that HL should be described as a shared responsibility between the patients and HCP. Attention should be focused to the HCP’s responsibility to adapt the communication to the patient’s knowledgebase.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Duo to the sensitive nature of data in this study, patients and HCP were assured raw data would remain confidential and would not be shared.

Ethics statements

Patient consent for publication

Ethics approval

The study was approved by the Privacy Ombudsman and the Hospital Investigational Review Board March 08 2019, reference number 2019/6465. Participants gave informed consent to participate in the study before taking part.


Supplementary materials


  • KRB and SER are joint first authors.

  • Contributors SER, SKS, MM and LM conceptualised the study and developed the method. KRB, HBL and YA contributed to development of the method. KRB, SER and HBL conducted the data collection. KRB, SER, SKS, HBL, YA and LM analysed and interpreted the patient data. KRB, SER and LM wrote the original draft. YA, HBL, SKS and MM were major contributors to the writing, review and editing. LM is acting as guarantor for the overall content. All authors read and approved the final manuscript.

  • Funding The study has been funded by the Hospital Pharmacies Enterprise, South Eastern Norway and the University of Oslo. Award/Grant number is not applicable.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.