Use and non‐use of a nationwide patient portal – a survey among pharmacy customers

A nationwide patient portal (My Kanta) for viewing electronic prescriptions and health data has been phased in since 2010 in Finland. This study aimed to study how commonly Finnish pharmacy customers use My Kanta, the factors related to My Kanta use, the main reasons for non‐use and how non‐users would like to monitor their medication and health information.

In Finland, a nationwide patient portal (My Kanta) and e-prescriptions were phased in by law since 2010. [20][21][22] These form part of the nationwide digital health care and social welfare Kanta services for health care professionals and the public. [13,20] My Kanta is available for everyone with a Finnish identity number and an ID for electronic services, such as online banking codes. [23] This portal includes information on all prescriptions and health data recorded in both public and private health care (Table 1). [23] Two thirds of Finnish adults use My Kanta. [24] The aim of My Kanta is to enhance patients' active involvement and increase their opportunities to take care of their medication and health. [25] Five years after the introduction of this portal, a study showed that a substantial proportion of pharmacy customers were unfamiliar with My Kanta; of those who knew the service, one in four disregarded it for unknown reasons. [26] The aim of this study was to investigate how commonly My Kanta is used by pharmacy customers, the factors related to use of the service, the main reasons for non-use and how non-users would like to monitor their medication and health information.

Study setting
In spring 2019, 18 community pharmacies in six regions across mainland Finland handed out questionnaires. [27] One University Pharmacy branch, one big city pharmacy and one small rural pharmacy were recruited from each region using convenience sampling. The number of questionnaires supplied to each pharmacy was based on the number of prescriptions dispensed annually by the pharmacy and varied between 40 and 320. The study focused on customers 18 or older purchasing prescription medicines for themselves or for their child under 18. After dispensing prescription medicine, pharmacists were instructed to inform the customer about the study and offer a questionnaire. Customers filled in the questionnaires at home and posted them in return envelopes to the research group. Pharmacies distributed the questionnaires as long as they had forms for a maximum of two weeks. The research group was informed about the number of remaining questionnaires for response rate calculation. Pharmacies were not required to keep a record of customers refusing to participate. Anonymous recruitment meant no reminders could be sent. Pharmacies distributed 2866 questionnaires.

Questionnaire
The four-page questionnaire consisted of 22 structured, Likert-scale and open-ended questions. The form had three parts. The first part was for all respondents and concerned background information, the second part was for respondents who used My Kanta, and the third part for respondents who did not use My Kanta. The questionnaire was designed on the basis of previous studies and My Kanta pages. [1][2][3][4][5]23,26,28,29] It was tested for face validity by three researchers experienced in designing questionnaire surveys. The questionnaire and data collection procedure were then piloted at one pharmacy. After customers filled in the questionnaire, they were interviewed by researchers about the intelligibility of the questions, leading to minor revisions. This paper reports the results from the first and third parts of the questionnaire. The use of My Kanta was surveyed with a structured question: 'Do you use My Kanta to browse your prescription and/or health information'?. The response options were as follows: 'Yes, I do', 'I have used it, but I am not going to use it any more' and 'I have never used it'. The questions about reasons for non-use and ways to monitor information were open-ended: 'What are the main reasons why you do not use My Kanta'? and 'How would you like to monitor your prescription and health information'?. Background information concerned demographics (gender, age, education, region), internet use, use of the internet to search for health-related information, and whether the respondent had an ID for electronic services, any chronic diseases diagnosed by a physician, and regularly used prescription medicines. Background information was obtained by means of structured questions, except for two open-ended questions (age and number of regularly used prescription medicines).

Data analysis
The data analysis included quantitative and qualitative phases. In the quantitative phase, differences between respondents and relationships between respondent characteristics and the use of My Kanta were examined using the Chi-square test, Fisher's exact test and t-test. Statistical significance was determined as P < 0.05. IBM SPSS Statistics Viewing and removing personal well-being data recorded via wellbeing applications • Weight, steps and activity during the day.
Giving consent to or limiting disclosure of personal data Saving a living will and organ donation testament Acting on behalf of dependants under 10 years of age • Viewing e-prescriptions and health data.
• Requesting a prescription renewal.
• Consenting to disclosure of dependant's data.
for Windows (Version 25.0 SPSS Inc., Chicago, IL, USA) was used in the analyses. Bivariate logistic regression analysis was used to determine the adjusted association between respondent characteristics and My Kanta use. Logistic regression analysis was conducted among respondents who had means to use the service, that is, the respondent used the internet and had an ID for electronic services (see Introduction). Thus, respondents who responded that they do not use the internet or do not have an ID for electronic services were excluded from the analysis. In addition, the respondents who did not know whether they have an ID for electronic services or who had a missing response regarding the use of internet or having an ID for electronic services were excluded. In the analysis, My Kanta use was compared with non-use of the service; response options 'I have used it, but I am not going to use it any more' and 'I have never used it' were combined into non-use. Covariates in the analysis were gender, age, education, region, internet use, use of the internet to search for health-related information, existence of any chronic disease and the number of currently used regular prescription medicines. Age was classified into four groups: 18-34, 35-59, 60-74 and ≥75. The response 'Don't know' concerning chronic diseases diagnosed by a physician was considered a missing value. The number of currently used prescription medicines served as a continuous covariate. The results of the analysis are expressed as odds ratios (ORs) with 95% confidence intervals (95% CIs).
In the qualitative phase, responses to the open-ended questions were encoded and categorised using inductive content analysis. [30] The responses were initially stored in Word 2016 (Microsoft Corporation, Redmond, WA, USA) and studied to provide familiarity with the data as a whole.
An analysis unit could be a word, a sentence, or a group of sentences describing a point related to the question. A response containing more than one such point was separated into several analysis units. These units were then simplified and sorted into emerging subcategories, which were named based on content. Similar subcategories were unified into main categories, and the main categories also were named based on content. One researcher (MS) conducted the inductive content analysis but continuously discussed it with the research group. Finally, the data were saved into SPSS and analysed quantitatively using frequencies.

Ethical statement
According to the national ethical instructions for research, [31] this survey does not require ethical approval. However, approval was conducted by the Committee on Research Ethics of the University of Eastern Finland on the request of the funding organisation (statement 23/2018). Participation in this study was voluntary; answering the questionnaire and posting it to the research group was regarded as informed consent to participate. Pharmacy owners gave their consent to distribute questionnaires at their pharmacies.

Study population
In total, 996 questionnaires were returned, two of which were blank ( Figure 1). Consequently, 994 questionnaires (34.7%) were included in the study. Most respondents (69.4%) were female ( Table 2). Respondents' ages ranged from 18 to 99 years (mean 62 years and median 66 years). Of all respondents, 92.0% had the means to use My Kanta, and they were younger, more educated, and had a chronic disease less often than respondents who did not have the means.
Factors related to the use of My Kanta among respondents with the means to use the service Most respondents who had the means to use My Kanta used the service (88.9%) ( Table 2). Users and non-users differed Table 2 Characteristics and differences between respondents who had and who did not have the means to use My Kanta Respondents with means 2 to use the service (n = 914) n (%) Respondents without means 2 to use the service (n = 71) n (%)

The main reasons for non-use of My Kanta
Almost all non-users (94.8%) reported at least one reason for not using the service. Non-users with the means most commonly stated that they did not need to use the service (60.4%) ( Table 4). They also reported having difficulty using the service (22.9%) or were unfamiliar with it (16.7%). Non-users without the means said they did not use the service because they lacked the tools (55.2%). Respondents in this group also had difficulty using the service (17.9%) or were unfamiliar with it (14.9%).
Respondents who stated they had no need to use My Kanta reported having no information to monitor, monitoring their information in other ways or experiencing no need to use it. Difficulty using the service included the inability to use a computer, the internet or My Kanta. Unfamiliarity with the service included reasons such as not knowing the service at all or having heard about it but not having familiarised oneself with it. Lack of tools included reasons such as not having or using a computer, smartphone, internet or ID for electronic services. Some non-users had the tools, but either these were not working or they did not want to use them.

How to monitor prescription and health information
Of all non-users of My Kanta, most (68.4%) reported at least one way they would like to monitor their prescription and health information. Almost a third (29.3%) of non-users with the means to use My Kanta reported that they would like to monitor information via My Kanta (Table 5). They explained that they should familiarise themselves with the service, they would like to learn to use it, or they would use the service when necessary. A pharmacy (26.7%) and physician/health centre (24.0%) were also commonly reported information sources. Non-users without the means  1 Respondents who had the means to use the service were included in the analysis (n = 914), of whom 105 were excluded due to the missing covariate values. 2 Per additional medicine.

Discussion
Most (82.5%) Finnish pharmacy customers surveyed used My Kanta. The use of service has increased significantly since 2015 (45.7%) when My Kanta use among pharmacy customers was last studied. [26] In the present study, My Kanta use was more common than among the Finnish adult population in general (63%). [24] This may result from selecting adult pharmacy customers purchasing prescription medicines as the study population because these were potential My Kanta users. In addition, users of the service may have been more interested in participating in the study. Correspondingly, internet use and having online banking IDs were more common in the present study than in the Finnish population in general; according to the Official Statistics of Finland, 89% of adults aged 16-89 use the internet and 83% use online banking (i.e. have an ID for eservices), [32] whereas in the present study, the corresponding proportions were 94% and 94%, respectively.
Searching for health-related information on the internet was the strongest predictor for My Kanta use. Logically, people who search for health-related information on the internet also use patient portals because these provide access to information about their own medicines and health. The association between searching for health information on the internet and portal use has been rarely studied; however, in one study, searching the internet for information about one's own illness was associated with patient portal use. [5] In contrast, the association between health literacy and portal use has been more studied, although with conflicting results. [4,15,18] Studies are needed to determine whether those who use patient portals have adequate health literacy skills to enable them to understand the information provided by these services.
The greater the number of regularly used prescription medicines, the more likely customers were to use My Kanta. In a previous Finnish study, pharmacy customers said that My Kanta provides a good summary of their prescribed medication and facilitates the monitoring of e-prescription information. [26] Understandably, an increasing number of prescriptions makes keeping up-to-date with one's medication more difficult, leading to an increased need to use the service. Furthermore, renewal requests might increase use of the service among regular users of prescription medicines. [33] In 2018, 2.1 million renewal requests were made by patients via My Kanta. [24] The present study showed some regional differences in using My Kanta. People living in Northern and Eastern parts of Finland were more likely My Kanta users. This result is partly in line with the statistics of Kanta services according to which My Kanta is mostly used in Northern, Southern, and Eastern parts of Finland. [24,34] Reasons for regional differences remain unclear but, in the present study, may partly result from regional differences in pharmacists' activity in delivering questionnaires. In addition, morbidity in Finland is the highest in Northern and Eastern parts of Finland. [35] There may also be regional differences in how actively healthcare units and pharmacies have informed patients about My Kanta and its use. [34] In contrast to previous studies about patient portals, [1,2,5,15,17,18,36] the present study showed that gender, age, education and the existence of chronic diseases were not associated with My Kanta use. However, the analysis of factors related to My Kanta use was conducted among respondents who had the means to use the service. Respondents without the means, who were older, less educated and chronically ill, were therefore excluded from the analysis. The digital divide, that is disparities in access to technology, is a widely known challenge of digitalisation. [1] People unable to use information technology must be taken into consideration when developing digital services [37] ; they need alternative ways to monitor health and medication information. Oral information from healthcare professionals may be enough for some people, whereas printed summaries about e-prescriptions and health information may be necessary for others. Healthcare professionals have an important role to play in identifying these customers so that the necessary information can be accurately targeted. Currently, healthcare units can print out information about health data and e-prescriptions to patients from patient data systems, but printing practices vary between units. Developing congruent practices to print corresponding information as in My Kanta for patients who do not use My Kanta might ensure patients' equality to get written information.
Around 10% of customers who had the means to use My Kanta did not use it, but still reported My Kanta as their preferred way to monitor their health and medication information. They most commonly reported that they had no need to use the service. However, it is unclear whether these people really were without the need or whether they were unaware that they should monitor their medication and health information. Patients perceiving themselves as subordinates to healthcare professionals may be a major barrier for patient engagement and involvement in their own care. [12] It is therefore important that patients know the purpose of patient portals and are encouraged to participate actively in their own health care. Difficulty using the service and unfamiliarity with it were also reported as barriers to My Kanta use, which is in line with previous studies. [2,3,5,19] Guidance is required in order to increase the use of My Kanta. The Kanta services have produced an online course about My Kanta, [38] and patients who find it difficult to use could be better informed about this course by healthcare professionals. In addition, a mobile application could encourage people to sign into the portal and increase its use because mobile phones are the most commonly used device for internet use in Finland. [32] Computers at pharmacies or healthcare centres may cater for those who need assistance with the use of a computer and My Kanta. Although unfamiliarity with My Kanta has notably decreased since 2015 when 37.9% of pharmacy customers did not know the service at all, [26] the present study suggests that the information given about the service has not reached all citizens and that its dissemination should continue.
The present study has both strengths and limitations. The study sample was large and included pharmacy customers across Finland. The goal of reaching both My Kanta users and non-users was also achieved. However, the response rate was lower than in surveys conducted with the same method earlier (40%À44%). [26,39] The study lacked information about the customers refusing to participate in the study. Therefore, the real response rate may be even lower than reported. There was also no information about the customers not returning the questionnaire. In addition, comparable statistics about adult pharmacy customers' characteristics are lacking. Compared to those who have received reimbursement for medicine costs under the Health Insurance Scheme, the respondents were older and more commonly women. [40] However, the study population was similar to that used in surveys among pharmacy customers conducted using the same method about generic substitution and e-prescriptions in Finland. [26,39] The questionnaire lacked any validated measures, but some questions were the same as in previous studies. [2,3,26,28,29] Conclusions Most Finnish pharmacy customers use My Kanta. Searching for health-related information on the internet and active internet use are the strongest predictors for use of the service. However, some customers do not use My Kanta although they have the means to do so. The most common reason for non-use is a lack of need. Guidance and information about the service and its benefits are needed in order to encourage more people to adopt it. Some customers are unable to use My Kanta and want to monitor their medication and health information via healthcare professionals. To establish equality with users of the service, these customers need to be identified and given oral and written information about their medication and health.

Conflict of interest
The Author(s) declare(s) that they have no conflicts of interest to disclose.

Funding
This work was supported by The Social Insurance Institution of Finland (Kela) (23/26/2018