This paper is in the following e-collection/theme issue:

Adoption and Change Management of eHealth Systems (670) Demographics of Users, Social & Digital Divide (677) Clinical Communication, Electronic Consultation and Telehealth (542) eHealth Infrastructures (306) Information Seeking, Information Needs (419) JMIR Theme Issue: COVID-19 Special Issue (2388)

Published on in Vol 10, No 4 (2022): April

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/37500, first published .
The Factors Associated With Nonuse of and Dissatisfaction With the National Patient Portal in Finland in the Era of COVID-19: Population-Based Cross-sectional Survey

The Factors Associated With Nonuse of and Dissatisfaction With the National Patient Portal in Finland in the Era of COVID-19: Population-Based Cross-sectional Survey

The Factors Associated With Nonuse of and Dissatisfaction With the National Patient Portal in Finland in the Era of COVID-19: Population-Based Cross-sectional Survey

Authors of this article:

Emma Kainiemi1 Author Orcid Image ;   Tuulikki Vehko1 Author Orcid Image ;   Maiju Kyytsönen1 Author Orcid Image ;   Iiris Hörhammer2 Author Orcid Image ;   Sari Kujala3 Author Orcid Image ;   Vesa Jormanainen1 Author Orcid Image ;   Tarja Heponiemi1 Author Orcid Image
Article Authors Cited by (6) Tweetations (2) Metrics

    Original Paper

    1Finnish Institute for Health and Welfare, Helsinki, Finland

    2Department of Industrial Engineering and Management, Aalto University, Espoo, Finland

    3Department of Computer Science, Aalto University, Espoo, Finland

    Corresponding Author:

    Emma Kainiemi, RM, MNSc

    Finnish Institute for Health and Welfare

    PO Box 30

    Mannerheimintie 166

    Helsinki, 00271

    Finland

    Phone: 358 295248292

    Email: emma.kainiemi@thl.fi


    Background: In the abnormal circumstances caused by the COVID-19 pandemic, patient portals have supported patient empowerment and engagement by providing patients with access to their health care documents and medical information. However, the potential benefits of patient portals cannot be utilized unless the patients accept and use the services. Disparities in the use of patient portals may exacerbate the already existing inequalities in health care access and health outcomes, possibly increasing the digital inequality in societies.

    Objective: The aim of this study is to examine the factors associated with nonuse of and dissatisfaction with the Finnish nationwide patient portal My Kanta Pages among the users of health care services during the COVID-19 outbreak. Several factors related to sociodemographic characteristics, health, and the use of health care services; experiences of guidance concerning electronic services; and digital skills and attitudes were evaluated.

    Methods: A national population survey was sent using stratified sampling to 13,200 Finnish residents who had reached the age of 20 years. Data were collected from September 2020 to February 2021 during the COVID-19 pandemic. Respondents who had used health care services and the internet for transactions or for searching for information in the past 12 months were included in the analyses. Bivariate logistic regression analyses were used to examine the adjusted associations of respondent characteristics with the nonuse of My Kanta Pages and dissatisfaction with the service. The inverse probability weighting (IPW) method was applied in all statistical analyses to correct for bias.

    Results: In total, 3919 (64.9%) of 6034 respondents were included in the study. Most respondents (3330/3919, 85.0%) used My Kanta Pages, and 2841 (85.3%) of them were satisfied. Nonusers (589/3919, 15%) were a minority among all respondents, and only 489 (14.7%) of the 3330 users were dissatisfied with the service. Especially patients without a long-term illness (odds ratio [OR] 2.14, 95% CI 1.48-3.10), those who were not referred to electronic health care services by a professional (OR 2.51, 95% CI 1.70-3.71), and those in need of guidance using online social and health care services (OR 2.26, 95% CI 1.41-3.65) were more likely nonusers of the patient portal. Perceptions of poor health (OR 2.10, 95% CI 1.51-2.93) and security concerns (OR 1.87, 95% CI 1.33-2.62) were associated with dissatisfaction with the service.

    Conclusions: Patients without long-term illnesses, those not referred to electronic health care services, and those in need of guidance on the use of online social and health care services seemed to be more likely nonusers of the Finnish nationwide patient portal. Moreover, poor health and security concerns appeared to be associated with dissatisfaction with the service. Interventions to promote referral to electronic health care services by professionals are needed. Attention should be targeted to information security of the service and promotion of the public’s confidence in the protection of their confidential data.

    JMIR Med Inform 2022;10(4):e37500

    doi:10.2196/37500

    Keywords

    patient portal; digital technology; eHealth; nonuse; dissatisfaction; COVID-19; digital health; patient empowerment; epidemiology; population survey; national survey 


    The worldwide COVID-19 pandemic limited the provision of nonurgent health care services [1,2]. During this time, the use and interest in patient portals increased [3,4] because portals have enabled patients to have continuous [5] and secure access to their health care documents and medical information [4]. Patient portals are electronic services that allow patients to access [6] and in some cases manage their electronic health record documentations [7] and interact with health care professionals [6,8,9]. The functionalities provided in a patient portal vary by portal and country [9,10].

    Patient portals offer transparent information about the patients’ health and well-being [11] and enhance the delivery of individualized care [3]. Patient portals have been reported to increase the patients’ knowledge and understanding of their own health condition, and thus they might be better prepared for future contacts with health care professionals [12]. This supports patient empowerment and engagement [4,5] as the patients feel more involved and responsible for their own care [12]. In addition, patient portals might increase the patients’ satisfaction with care [13-18] and improve patient safety [15,17]. However, high-quality evidence of health benefits has not yet been demonstrated [8].

    The potential benefits of patient portals cannot be utilized unless the patients accept and adopt the service [7,19]. According to the information system (IS) success model, the benefits of using the service arise from its use and user satisfaction [20]. Further, increased user satisfaction will lead to increased use [20], and unmet expectations will alter the use and satisfaction with patient portals [21]. Not all the barriers related to the use of patient portals are related to practical issues, such as a lack of hardware and access to the internet, but patients may have other valid reasons for nonuse as well [7,22,23]. Patients are also in an unequal position in terms of using electronic health care services, since not everyone has the resources or the same possibilities to use the services and take more responsibility for the management of their own health and well-being [24].

    Previous research has examined differences in patient portal use in different contexts and patient populations. Several studies have reported an association between portal use and sociodemographic background [6,7,14,23,25,26] and various health-related factors [7,12,14,25,27,28]. In addition, patients’ guidance through increasing awareness and knowledge of patient portals [6,25,28], as well as endorsement and engagement with portals by health care professionals [7,25], have been identified as important associated factors. There are also some studies that have reported an association between the use of patient portals and factors related to the use of the internet, such as the frequency of use [26,29,30] and perceptions of the users’ own internet skills [26,30]. Furthermore, it has been reported that perceptions of electronic services may encourage or impede their use [31]. However, these previous studies have been conducted under normal circumstances before the COVID-19 pandemic and are thus only partially applicable in the context changed by the pandemic [4].

    Factors associated with the patients’ satisfaction with patient portals have been less studied. Mainly descriptive research on the portal users’ experiences exists, and only little research has been conducted with quantitative methods about factors associated with satisfaction [5]. Kong et al [5] examined factors that predicted portal use and the users’ willingness to recommend the service among chronically ill patients during the COVID-19 pandemic in the Netherlands. They discovered that the respondent’s level of control, hospital visit time, life satisfaction, and level of depression are significantly associated with portal use. Variables related to the portal user’s waiting times for responses via the portal were the strongest predictors of the willingness to recommend the portal. However, the used variables concerned patients with long-term illnesses, and no comparisons were made to assess whether the same variables were associated with the use and willingness to recommend the patient portal. In addition, only little research exists on the factors associated with use and satisfaction using a nationally representative sample.

    The aim of this study is to examine factors associated with the nonuse of and dissatisfaction with the Finnish nationwide patient portal My Kanta Pages (My Kanta) during the COVID-19 pandemic. Only respondents who had used the internet in the past 12 months were included to examine nonuse beyond the first-level digital divide [32] caused by a lack of hardware and access to the internet. Several factors related to (1) sociodemographic characteristics, (2) health and the use of health care services, (3) experiences of guidance concerning electronic services, and (4) digital skills and attitudes were examined. The evaluation of factors associated with the nonuse of and dissatisfaction with the patient portal is important to further develop the service and advocate for nonusers. Disparities in the use of patient portals might exacerbate the already existing disparities in health care access and health outcomes [33], increasing digital inequality in societies [34]. Knowledge of the factors that are associated with the nonuse of and dissatisfaction with the national patient portal in Finland, one of the pioneer countries in digitalization, can provide valuable information for countries and organizations that are further developing their electronic services.


    Study Context

    Finland is a sparsely populated country with 5.5 million residents. The health care system is decentralized, and until the end of 2022, municipalities (n=311) are responsible for organizing health care services, which are funded by taxes, state transfers, and user fees [1]. Finland can be considered 1 of the leading countries in terms of digitalization [35]. One of the most widely used electronic service is the nationwide patient portal called My Kanta. Between the years 2010 and 2018, cumulatively 63% of the adult Finnish residents had accessed the service. There is a professional user interface for Kanta Services, which can be used by the public and private actors of the social welfare and health care sector [36]. In addition to the nationwide patient portal, some public and private actors offer their clients access to their own patient portals or regional portals [37].

    My Kanta was launched step-by-step starting from 2010 [38] to promote patient safety as well as the continuity and transparency of care [39]. My Kanta enables continuous access of Finnish residents to their health information [9], including browsing their own electronic prescriptions and medical records, such as patient reports, laboratory results, and X-ray statements [40]. All producers of health care services have been obligated to use electronic prescriptions since 2017 [41], and patients can request prescription renewals via My Kanta [42]. In addition, it is possible to record and monitor well-being data, such as blood glucose or activity meter. To access My Kanta, a Finnish personal identity code is needed, and e-authorization must be made with identification using online banking codes, mobile identification, or a certificate card [42].

    The number of My Kanta users has grown steadily since its launch [38], and the COVID-19 pandemic increased the number of logins because of the availability of coronavirus test results [43]. In 2020, the service was used 29.4 million times by a total of 2.7 million individual visitors [43]. E-prescriptions were issued approximately 26.4 million times during 2020 [44]. In the future, the use is expected to further increase as the deployment of authorization for an adult to act on behalf of another adult was introduced after the data collection period and authorization for guardians to act, with some restrictions, on behalf of their children aged 10-17 years [45] will be fully implemented.

    Sample

    This study was conducted in Finland as part of the FinSote 2020 National Survey of Health, Wellbeing, and Service Use [46]. The questionnaire was sent using stratified sampling to 13,200 Finnish residents who had reached the age of 20 years. Data were collected from September 2020 to February 2021 during the second wave of the COVID-19 pandemic. A possibility to respond either in electronic or in paper form in Finnish, Swedish, Russian, or English was offered. During the data collection, participants who had not responded were approached by mail up to 4 times.

    Altogether, 6034 Finnish residents (n=3401 [56.4%] female, mean age 64.5 years, SD 17.9) responded to the questionnaire (response rate 46.5%). In total, 3919 (65.0%) respondents were included in the study sample as they had used health care services and the internet in the past 12 months. The sample was weighted using inverse probability weighting (IPW) correction [47]. The weights were estimated using sociodemographic register–based variables: the respondents’ age, gender, marital status, level of education, area of residence, and native language. Information about the respondents’ age, gender, and area of residence were obtained from the National Population Register. In previous research, the IPW method improved the accuracy of the results of a population survey and removed most of the bias caused by nonresponse in the various subpopulations [48].

    Ethics Approval

    Participation in the study was completely voluntary. Ethical approval was obtained from the Ethics Committee of the Finnish Institute for Health and Welfare (THL/637/6.02.01/2017).

    Measurements

    Dependent Variables

    The nonuse of My Kanta was evaluated with the question “Have you used My Kanta in the past 12 months?” Respondents were asked to respond (1) no or (2) yes. For the analyses, the measure was binary-coded (0=user, 1=nonuser), and the users of My Kanta were set as the reference group.

    The dissatisfaction with My Kanta was evaluated with a question concerning satisfaction: “If you have used the service, assess the quality of the service using a school grade (4-10).” In the Finnish education system, grades 8-10 represent grades from good to excellent and grades 4-7 from fail to satisfactory [49]. For the analyses, the measure was binary-coded (0=respondent was satisfied [grades 8-10] and 1=respondent was dissatisfied [grades 4-7] with the service), and the satisfied users of My Kanta were set as the reference group. Because the research interest was in respondents who were less satisfied with the service, respondents who gave an assessment of grade 7 were included in the group of dissatisfied users. This decision was also made based on substantive judgment to even the distribution [50] between satisfied and dissatisfied users, since only a small number of respondents had selected a grade from 4 to 6.

    Independent Variables

    Independent variables included characteristics concerning (1) sociodemographic background, (2) health and the use of health care services, (3) experiences of guidance concerning electronic services, and (4) digital skills and attitudes. All the used variables are presented in Multimedia Appendix 1.

    Sociodemographic Characteristics

    The respondents’ sociodemographic characteristics included their age, gender, education, and degree of urbanization. Age was used as a categorical variable in the descriptive statistics and as a continuous variable in all analyses. The degree of urbanization was determined according to the municipal classification and divided into 3 categories according to the proportion of people living in urban settlements and the population of the largest urban settlement: urban, semiurban, and rural municipalities [51]. Because of age-related differences in education, the respondents’ educational level was first divided into 10-year age groups by gender. Each group was divided into 3 categories based on their years of education, with approximately one-third of the respondents in each category: low, median, and high. Hence, the education-level variable had hardly any interaction with age and gender.

    Health and the Use of Health Care Services

    Variables concerning the respondents’ health and the use of health care services included self-rated health, long-term illness, and the use of health care services. Self-rated health was evaluated with a widely used, single-item measure of self-perceived health status. A subjective assessment of own health has been reported as a more sensitive measure in health monitoring than external measures of health, since it includes biological, psychological, and social dimensions. [52]. A scale from good to poor was used to evaluate the present state of the respondents’ health. In the analyses, the options (1) good and (2) fairly good were combined to represent good health, and the remaining options represented average or poor health. Long-term illness was binary-coded as (1) yes and (2) no. The use of health care services was binary-coded according to the number of annual outpatient appointments with a physician; 8 or more annual appointments were considered a high use of health care services, and less than 8 were counted as low or average use [53].

    Experiences of Guidance Concerning Electronic Services

    Variables concerning the experiences of guidance concerning electronic services included referrals to electronic services and the need for guidance on how to use online social or health care services. The referral to electronic services was evaluated with the question “If you have used social or health care services in the traditional way (paper, visit, or call) in the past 12 months, were you referred to electronic services (eg, My Kanta)?” For the analyses, option (1) yes, I was referred represented the respondents who were referred to electronic health care services. Option (2) was for those who were not referred to electronic health care services.

    The need for guidance on using online social and health care services was evaluated with the statement “I need help with using online social and health care services.” In the analyses, the options (1) completely agree and (2) somewhat agree were combined as (1) yes and the remaining options as (2) no or no opinion.

    Variables Related to Digital Skills and Attitudes

    Variables related to digital skills and attitudes included digital skills, perceived benefits of electronic social and health care services, and security concerns. Digital skills were evaluated with 6 validated statements [54]. Based on pilot testing, 2 (33%) of the statements were transferred into positive statements [37]. A 5-point Likert scale was used to answer the statements (1=completely agree to 5=strongly disagree). Cronbach α for the statements was .86. In the analyses, a mean variable ranging from 1 to 5 was calculated for each respondent, and the measure was binary-coded to indicate (1) good skills (mean≤2.5) and (2) poor skills (mean>2.6). The same coding has previously been used in national research [37].

    The perceived benefits of electronic social and health care services were measured with 8 statements. A 5-point Likert scale was used to answer the statements (1=completely agree to 5=strongly disagree). Cronbach α for the statements was .91. In the analyses, missing values were coded as neither agree nor disagree. A mean variable from 1 to 5 was calculated for each respondent, and the measure was binary-coded as (1) beneficial (mean≤2.5) and (2) unbeneficial (mean>2.6). The same coding has previously been used in national research [37].

    Security concerns were evaluated with the statement “I am concerned about information security when it comes to my personal details”. In the analyses, options (1) completely agree and (2) somewhat agree were combined as (1) yes and the remaining options as (2) no.

    Statistical Analysis

    In all statistical analyses, the IPW method [47] was applied to correct for bias by handling both differential sampling probabilities and missing data. Due to nonresponse in some items, the number of observations varied in the analyses.

    Bivariate logistic regression analyses were used to examine the adjusted associations of respondent characteristics with the nonuse of My Kanta and dissatisfaction with the service (in separate analyses). First, univariate analyses, adjusted for age, gender, and education, were conducted at a time to examine the association of the dependent variable with each independent variable. Second, a multivariable model was formed, including only those independent variables with a P value of <.10. This cut-off for the P value was used for including the variables in the multivariable model, because the purpose was to identify potential independent variables rather than to test a hypothesis [55]. In the fully adjusted multivariable model, a P value of <.05 was considered statistically significant. Statistical methods suitable for weighted data were used, and SPSS Statistics version 27 was applied for the analyses.


    Characteristics

    The weighted majority (3330/3919, 85.0%) of the respondents had used My Kanta in the past 12 months. Most of the My Kanta users (2841/3330, 85.3%) were satisfied with the service. A minority of respondents (589/3919, 15%) had not used My Kanta in the past 12 months.

    The IPW weighted characteristics of the respondents representative of the Finnish population are presented in Tables 1-4. Almost half of the respondents were aged between 35 and 59 years. Over half (n=3401, 56.4%) of the respondents were female, and the majority lived in urban regions. Over half of the respondents were not referred to electronic health care services, such as My Kanta, by a health care professional. About half of the respondents perceived electronic health care services to be beneficial. The respondents who had used My Kanta in the past 12 months were mostly satisfied with the service (mean 8.31, SE .03), whereas a minority of users (489/3330, 14.7%) were dissatisfied. Over one-third (1359/3919, 34.7%) of the respondents had also used an electronic service provided by their occupational health care provider. Of these respondents, a minority (130/1359, 9.6%) only used the service provided by their occupational health care provider and not the nationwide patient portal My Kanta.

    Table 1. Sociodemographic characteristics of the weighted study samplea.
    CharacteristicsRespondents, n (%)Nonusers (N=589), n (%)Dissatisfied users (N=489), n (%)
    Age, years (N=3919)

    		20-34918 (23.4)144 (24.5)120 (24.5)

    		35-591773 (45.2)286 (48.5)234 (48.0)

    		60-741008 (25.7)129 (21.9)107 (21.8)

    		75-99220 (5.6)30 (5.1)28 (5.7)
    Gender (N=3919)

    		Male1641 (41.9)301 (51.1)221 (45.2)

    		Female2278 (58.1)288 (48.9)268 (54.8)
    Education (N=3873)

    		Low1499 (38.7)232 (39.3)196 (40.1)

    		Median1254 (32.4)187 (31.8)126 (25.8)

    		High1120 (28.9)170 (28.9)167 (34.1)
    Degree of urbanization (N=3919)

    		Urban2918 (74.5)427 (72.5)373 (76.4)

    		Semiurban536 (13.7)78 (13.3)64 (13.1)

    		Rural465 (11.9)84 (14.2) 52 (10.6)

    aInverse probability weighting (IPW)-corrected.

    Table 2. Health and service use by the weighted study samplea.
    CharacteristicsRespondents, n (%)Nonusers (N=589), n (%)Dissatisfied users (N=489), n (%)
    Self-rated health (N=3893)

    		Average or poor1251 (32.1)133 (22.6)245 (50.2)

    		Good2642 (67.9)456 (77.4)244 (49.8)
    Long-term illness (N=3857)

    		Yes2165 (56.1)191 (32.4)319 (65.2)

    		No1692 (43.9)398 (67.6)170 (34.8)
    Use of health care services (N=3835)

    		Low or average3516 (91.7)580 (98.5)435 (89.0)

    		High319 (8.3)9 (1.5)54 (11.0)

    aInverse probability weighting (IPW)-corrected.

    Table 3. Experiences of guidance concerning electronic services for the weighted study samplea.
    CharacteristicsRespondents, n (%)Nonusers (N=589), n (%)Dissatisfied users (N=489), n (%)
    Referral to electronic services (N=3166)

    		Yes1386 (43.8)154 (26.2)216 (44.1)

    		No1780 (56.2)435 (73.8)273 (55.9)
    Need for guidance (N=3833)

    		Yes379 (9.9)86 (14.6)74 (15.1)

    		No3454 (90.1)503 (85.4)415 (84.9)

    aInverse probability weighting (IPW)-corrected.

    Table 4. Variables related to digital skills and attitudes of the weighted study samplea.
    CharacteristicsRespondents, n (%)Nonusers (N=589), n (%)Dissatisfied users (N=489), n (%)
    Digital skills (N=3897)

    		Poor199 (5.1)48 (8.2)25 (5.1)

    		Good3698 (94.9)541 (91.8)464 (94.9)
    Perceived benefits (N=3919)

    		Yes1840 (47.0)252 (42.8)181 (37.1)

    		No2079 (53.0)337 (57.2)308 (62.9)
    Security concerns (N=3823)

    		Yes or N/Ab1923 (50.3)323 (54.9)312 (63.9)

    		No1900 (49.7)266 (45.1)177 (36.1)

    aInverse probability weighting (IPW)-corrected.

    bN/A: not applicable.

    Associations With the Nonuse of My Kanta

    Based on the results of age-, gender-, and education-adjusted univariate logistic regression analysis (Table 5), the following factors were included in the multivariable model: self-rated health, long-term illness, use of health care services, referral to electronic services, need for guidance, and digital skills.

    The results of the fully adjusted logistic regression analysis regarding the nonuse of My Kanta are presented in Table 6. Male respondents were more likely to be nonusers of My Kanta compared to females. Respondents who used health care services to a low or average degree and who did not have a long-term illness were more likely to be nonusers of My Kanta compared to those who used health care services to a high degree and had a long-term illness. In addition, respondents who were not referred to electronic services, needed guidance, or had poor digital skills were over 2 times more likely to be nonusers of My Kanta compared to their counterparts.

    Table 5. Results of univariate logistic regression analyses for the nonuse of My Kantaa,b.
    Univariate analyses characteristicscORd (95% CI)P valuee
    Sociodemographic characteristics

    		Age (years)0.84 (0.61-1.16).17

    		Gender (male)1.61 (1.21-2.12).001

    		Low educational level1.03 (0.73-1.46).85

    		Median educational level1.01 (0.71-1.42).99

    		High educational levelReferenceN/Af
    Degree of urbanization

    		UrbanReferenceN/A

    		Semiurban1.03 (0.70-1.52).87

    		Rural1.34 (0.93-1.92).12
    Health and service use

    		Self-rated health (good)1.75 (1.25-2.45).001

    		Long-term illness (no)3.24 (2.41-4.36)<.001

    		Use of health care services (low or average)6.75 (2.49-18.31)<.001
    Experiences of guidance concerning electronic services

    		Referral to electronic services (no)2.43 (1.70-3.49)<.001

    		Need for guidance (yes)2.09 (1.44-3.05)<.001
    Variables related to digital skills and attitudes

    		Digital skills (poor)2.37 (1.50-3.76)<.001

    		Perceived benefits (no)1.25 (0.94-1.67).13

    		Security concerns (yes)1.25 (0.94-1.67).13

    aInverse probability weighting (IPW)-corrected.

    bThe model included the main effect of each variable adjusted for age, gender, and education.

    cReference categories indicated in parentheses: gender: male vs female; self-rated health: good vs average or poor; long-term illness: no vs yes; use of health care services: low or average vs high; referral to electronic services: no vs yes; need for guidance: yes vs no; digital skills: poor vs good; perceived benefits: no vs yes; security concerns: yes vs no.

    dOR: odds ratio.

    eSignificance level of P<.10.

    fN/A: not applicable.

    Table 6. Results of the fully adjusted logistic regression analysis for the nonuse of My Kanta (N=2328)a.
    Multivariable model characteristicsbORc (95% CI)P valued
    Gender

    		FemaleReferenceN/Ae

    		Male1.67 (1.19–2.42).003
    Self-rated health

    		Average or poorReferenceN/A

    		Good1.48 (0.95–2.31).08
    Long-term illness

    		YesReferenceN/A

    		No2.14 (1.48–3.10)<.001
    Use of health care services

    		HighReferenceN/A

    		Low or average4.66 (1.29–16.84).02
    Referral to electronic services

    		YesReferenceN/A

    		No2.51 (1.70–3.71)<.001
    Digital skills

    		GoodReferenceN/A

    		Poor2.53 (1.32–4.83).01
    Need forguidance

    		NoReferenceN/A

    		Yes2.26 (1.41–3.65)<.001

    aInverse probability weighting (IPW)-corrected.

    bThe model included all the independent variables with a P value of <.10 in the univariate model adjusted for age, gender, and education.

    cOR: odds ratio.

    dSignificance level of P<.05.

    eN/A: not applicable.

    Associations With Dissatisfaction With the Use of My Kanta

    Based on the results of the age-, gender-, and education-adjusted univariate analyses (Table 7), the following variables were included in the multivariable model: education, self-rated health, long-term illness, need for guidance, perceived benefits, and security concerns. The results of the fully adjusted logistic regression analysis are presented in Table 8.

    In the fully adjusted multivariable model, respondents who were younger, were male, and had a high level of education were more likely to be dissatisfied with My Kanta compared to their counterparts. Respondents with average or poor self-rated health were over 2 times more likely to be dissatisfied with My Kanta compared to respondents with a good perception of their own health. In addition, respondents who perceived electronic services as unbeneficial, who needed guidance, and who had security concerns were more likely to be dissatisfied with My Kanta compared to their counterparts.

    Table 7. Results of univariate logistic regression analyses for dissatisfaction with My Kantaa,b.
    Univariate analyses characteristicscORd (95% CI)P valuee
    Sociodemographic characteristics

    		Age (years)0.99 (0.99–1.01).33

    		Gender (male)1.31 (0.95–1.82).09

    		Low educational levelReferenceN/Af

    		Median educational level0.72 (0.49–1.06).09

    		High educational level1.11 (0.76–1.62).58
    Degree of urbanization

    		RuralReferenceN/A

    		Semiurban1.04 (0.60–1.79).89

    		Urban1.08 (0.72–1.61).72
    Healthand service use

    		Self-rated health (average or poor)2.45 (1.79–3.36)<.001

    		Long-term illness (yes)1.34 (0.95–1.88).09

    		Use of health care services (high)1.29 (0.76–2.21).35
    Experiences of guidance concerning electronic services

    		Referral to electronic services (no)1.17 (0.82–1.67).38

    		Need for guidance (yes)2.98 (1.83–4.85)<.001
    Variables related to digital skills and attitudes

    		Digital skills (poor)1.63 (0.88–3.03).12

    		Perceived benefits (no)1.79 (1.30–2.47)<.001

    		Security concerns (yes)2.24 (1.61–3.13)<.001

    aInverse probability weighting (IPW)-corrected.

    bThe model included the main effect of each variable adjusted for age, gender, and education.

    cReference categories indicated in the parentheses: gender: male vs female; self-rated health: average or poor vs good; long-term illness: yes vs no; use of health care services: high vs average or low; referral to electronic services: no vs yes; need for guidance: yes vs no; digital skills: poor vs good; perceived benefits: no vs yes; security concerns: yes vs no.

    dOR: odds ratio.

    eSignificance level of P<.10.

    fN/A: not applicable.

    Table 8. Results of the fully adjusted logistic regression analysis for dissatisfaction with My Kanta (N=2341)a.
    Multivariable model characteristicsbORc (95% CI)P valued
    Age0.99 (0.98–1.00).002
    Gender

    		FemaleReferenceN/Ae

    		Male1.40 (1.00–1.94).05
    Education

    		LowReferenceN/A

    		High1.48 1.02–2.16 ().04
    Self-rated health

    		GoodReferenceN/A

    		Average or poor2.10 (1.51–2.93)<.001
    Long-term illness

    		NoReferenceN/A

    		Yes1.02 (0.71–1.45).92
    Perceived benefits

    		YesReferenceN/A

    		No1.52 (1.09–2.11).01
    Security concerns

    		NoReferenceN/A

    		Yes1.87 (1.33–2.62)<.001
    Need forguidance

    		NoReferenceN/A

    		Yes2.14 (1.33–3.46).002

    aInverse probability weighting (IPW)-corrected.

    bThe model included all the independent variables with a P value of <.10 in the model adjusted for age, gender, and education.

    cOR: odds ratio.

    dSignificance level of P<.05.

    eN/A: not applicable.


    Principal Results

    Most respondents of this nationally representative survey study had used the nationwide Finnish patient portal My Kanta in the previous 12 months and were satisfied with the service. However, more than every 10th user of health care services and the internet were nonusers of the national patient portal, and approximately the same number of users were dissatisfied with the service. Males and those in a need of guidance were more likely to be nonusers of the patient portal and dissatisfied with the service compared to women and those not needing guidance. Not having a long-term illness and low or average use of health care services were associated with the increased likelihood of nonuse of the My Kanta portal. In addition, respondents who were not referred to electronic services and who had poor digital skills were more likely to be nonusers of My Kanta compared to their counterparts. A younger age, higher education, and poor self-rated health were associated with an increased likelihood of dissatisfaction with the service. In addition, respondents who did not perceive electronic health care services to be beneficial and who had security concerns were more likely to be dissatisfied with the service compared to their counterparts.

    Strengths and Limitations

    Finland is 1 of the forerunners of digitalization and ranked highest in information exchange and patient-centered information processing in an international comparative study [56]. By presenting the characteristics that are associated with nonuse of and dissatisfaction with the nationwide patient portal in Finland, valuable information can be offered for national initiatives for improvement and other countries aspiring to provide their residents with access to their health care documentation. However, generalizing our findings to countries with different levels of digitalization or service system should be done with caution.

    A nationally presentative sample of Finnish residents was included in the analysis. The applied IPW method has previously been reported to improve the accuracy and generalizability of results [48]. However, the findings are based on self-reported data, with the possibility for bias, as respondents might not recall the previous happenings accurately. This could also lead to problems associated with common method variance and the inflation of the strength of relationships. In addition, some of the used independent variables are hard to explicitly measure and quantify because of their subjective nature, such as perceived benefits of electronic health care services. Although multiple factors were adjusted in the analyses, the possibility of residual confounding remains. Moreover, cross-sectional survey data do not allow drawing any confirmatory causal inferences from the results.

    Since only respondents who had used the internet for transactions or for searching for information were included in the study, the results are only applicable when considering the nonuse of patient portals beyond the first-level digital divide caused by the lack of necessary devices and access to the internet. Some respondents who did not use My Kanta used additional patient portals provided by private or public providers of health care services. It is also noteworthy that some respondents might not be referred to electronic health care services, because their transactions in health care do not require further action or electronic services cannot provide support in their situation. The research concerning the users and nonusers of nationwide patient portals is sparse, and comparison is difficult as the properties provided in the portals vary, in addition to the differing patient populations and adjustments in the analyses.

    Comparison With Prior Work

    The use of nationwide patient portals varies by country and portal [9,10]. This study showed that the majority of Finnish residents who had used health care services and the internet in the past 12 months had used the nationwide My Kanta patient portal. The use of the patient portal increased during the COVID-19 pandemic because of the availability of coronavirus test results, easing the burden on health care services [43]. The availability of the test results has likely increased the use of My Kanta among those with no long-term illness and low use of health care services. This may also suggest that a larger proportion of this group among the respondents was reached for this study than would have been reached before the pandemic. Approximately only every 10th user of the patient portal was dissatisfied, indicating a high level of satisfaction with the service. It might be presumed that the restrictions for avoiding face-to-face encounters and fear of the infection increased the overall satisfaction with electronic health care services during the COVID-19 pandemic. However, even before the pandemic, high satisfaction with My Kanta has been reported [57-59].

    The results of this study suggested that younger and more educated respondents were more likely to be dissatisfied with My Kanta compared to older and less educated respondents. The findings of this study were supported by the fact that younger generations have grown up with technology and were thus more comfortable using electronic services [60], which might result in higher expectations toward the services. Previous research on a Norwegian symptom checker [61] reported that compared to younger users, older users were more satisfied with the service because they tended to navigate it in a more superficial way without gaining awareness of the existing problems. Contradictory to the findings of this study, patients with higher education have been reported to be more satisfied with telemedicine compared to patients with lower education [62]. However, only participants with moderate or high levels of digital health literacy were recruited, which might explain the contradictory results. Among the less educated respondents, digital health literacy skills might be an important factor leading to dissatisfaction with the service [63].

    This study found that respondents without any long-term illness and with low or average use of health care services are more likely to be nonusers of My Kanta, which is consistent with previous research [7,19,25,27,64]. Patients without long-term illness and lower use of health care services might have fewer needs related to health care and the use of patient portals. In addition, good self-rated health was associated with patient portal nonuse in this study, similar to studies by Moll et al [27] and Zanaboni et al [12], but only before controlling for the use of health care services and long-term illness. However, even after these adjustments, a more negative perception of patients’ own health was associated with dissatisfaction with the service. Previous research has reported an almost linear association between the patients’ poor perception of their own health and the number of annual outpatient visits with a physician [52]. It can be anticipated that patients with poor self-rated health have more health care needs and fewer resources and, thus, presumably higher expectations of patient portals, which may lead to dissatisfaction.

    Over half of the respondents were not referred to electronic health care services by their care providers, and the nonreferred respondents were less likely to use the nationwide patient portal My Kanta compared to referred respondents. Sääskilahti et al [29] and Kong et al [5] have also reported unfamiliarity with the service among nonusers of patient portals. It is important to highlight the role of professionals and health care managers in activating and engaging patients to accept and use patient portals, because promotion is heavily associated with their use [7,29,65-69]. The promotion should be integrated into routine care processes, and individual training and support should be provided on the portal use to patients with different background demographics to help prevent the digital divide from widening [25,67,69-71]. In addition to the promotion by professionals, alternative means are also needed to increase adoption [70]. Further research is necessary to identify effective ways to integrate patient portal enrollment into clinical practice. After data collection, My Kanta received national publicity because of active marketing of the EU Digital COVID Certificate, which can be downloaded from the portal. This has further increased the public’s awareness of My Kanta. In the future, it might be expected that increasingly more patients have prior knowledge and experience with the service. This might also represent a significant incentive for the public’s further adoption of electronic services [4].

    Although patients have prior experience in the use of electronic services and information technology tools, the ability to review and manage medical records on patient portals should not be assumed [28]. Results of this study suggest that respondents who needed guidance in the use of electronic health care services were more likely to be nonusers of the portal or dissatisfied with the service compared to respondents without a need for guidance. It is likely that perception of a lack of ability in the use of electronic health care services or poorly designed services will alter the use and the benefits from their use, leading to dissatisfaction with the portal.

    The previous literature has suggested different ways in which the use of patient portals could be promoted, including ease of entry [13,30,72], easy navigation [73,74], and reducing the required cognitive demands [73]. Electronic services should meet certain accessibility requirements [75,76] in order to support the equality of the users. Many providers of electronic health care services still struggle with meeting the demands of accessibility, impairing the position of especially patients with disabilities [77]. My Kanta fulfills these requirements at a certain level by assessing and reporting the current status of the accessibility and providing an electronic channel for feedback [78].

    In addition to easy accessibility of the service, assistance on the use should be available at a low threshold. Because patients have previously been reported to seldom seek help from family members, friends, service support, or health care providers [12], electronic services should be designed to include assistance to users. New electronic introductory and teaching materials have been prepared by the system administrator of My Kanta [79], and health care professionals should guide their clients to these materials. Patients with different background demographics were involved to a limited extent in the development of these materials. Efforts to stimulate participation of especially disadvantaged patients in developmental work of patient portals [80] and educational materials should be highlighted.

    Digital skills are necessary for wider patient adoption and use of patient portals [31,34,37,81,82]. The findings of this study are similar to previous research [26,30] as respondents with poor digital skills were more likely to be nonusers of My Kanta portal compared to respondents with good skills. For the patients to be able to effectively navigate the portal, their digital competence needs to be promoted [71]. The Finnish national strategy for applying information technology to health care and social welfare currently states that Finnish residents should be able to use electronic services and produce self-recorded data to promote their well-being [83]. However, by making digital skills a policy priority, the equal use of electronic health care services and patient portals could be promoted and the risk for digital divide minimized [82]. Good digital skills have also been reported to be associated with the perception that electronic services are more useful [24].

    Attitudes about the usefulness, appropriateness, and potential downsides of electronic services may encourage or impede the use [31]. In this study, the respondents who did not perceive overall benefits in electronic health care services were more likely dissatisfied with My Kanta compared to respondents with a more positive attitude. Negative attitudes have previously been reported to alter patient satisfaction with the patient portal [69]. To ensure the widespread and equal use of electronic health care services, all users must experience them as beneficial [24]. Patients’ perceptions of the benefits can be increased by offering them demonstrations and information about the capabilities of the patient portal [69,84]. Perceived benefits were not associated with the nonuse of the patient portal according to the results of this study.

    According to this study, respondents who had security concerns were more likely to be dissatisfied with My Kanta compared to respondents who felt more secure. Similar to the results of Woods et al [30], security concerns were not associated with the use of the patient portal. Privacy, security, and confidentiality concerns regarding medical information have been identified as barriers to the use of patient portals, and some patients may feel discomfort at having their personal health information on the internet [25,66,69,85]. Attention needs to be paid to information security and identity protection, since these are critical issues and central to widespread consumer acceptance and adoption of patient portals [82]. Security concerns also complicate requests for assistance and guidance from non-health-care professionals as well as the use of patient portals on public computers, since others might be able to see sensitive medical information on the screen [69,84]. Private facilities should be promoted in libraries and other places with public computers. The COVID-19 pandemic has complicated the requests for assistance and the use of computers in public facilities because of societal restrictions. More research is needed to understand how safety concerns could be alleviated. In addition, technology users of all ages should be equipped with knowledge of online privacy and security as a new set of cyber security skills are needed in the increasingly digital society [86].

    Conclusion

    According to the results of this population-based cross-sectional survey study in the era of COVID-19, patients without long-term illnesses, those not referred to electronic health care services, and those in need of guidance on the use of online social and health care services seem to be more likely nonusers of the Finnish nationwide patient portal My Kanta. Moreover, poor health and security concerns seem to be associated with dissatisfaction with the service. Interventions to promote referral to electronic health care services by professionals are needed. Attention must be paid to information security of the service as well as the alleviation of the patients’ privacy concerns.

    Acknowledgments

    This study was supported by the Strategic Research Council at the Academy of Finland (projects 327145 and 327147), the Ministry of Social Affairs and Health (project 414919001), and NordForsk (project 100477). The authors wish to thank all the respondents for their contribution.

    Conflicts of Interest

    None declared.

    ‎
    Multimedia Appendix 1

    Description of the used variables.

    DOCX File , 16 KB
    1. Tiirinki H, Tynkkynen L, Sovala M, Atkins S, Koivusalo M, Rautiainen P, et al. COVID-19 pandemic in Finland: preliminary analysis on health system response and economic consequences. Health Policy Technol 2020 Dec;9(4):649-662 [FREE Full text] [CrossRef] [Medline]
    2. Ting D, Carin L, Dzau V, Wong TY. Digital technology and COVID-19. Nat Med 2020 Apr;26(4):459-461 [FREE Full text] [CrossRef] [Medline]
    3. Power K, McCrea Z, White M, Breen A, Dunleavy B, O'Donoghue S, et al. The development of an epilepsy electronic patient portal: facilitating both patient empowerment and remote clinician-patient interaction in a post-COVID-19 world. Epilepsia 2020 Sep;61(9):1894-1905 [FREE Full text] [CrossRef] [Medline]
    4. Stanimirovic D. eHealth patient portal: becoming an indispensable public health tool in the time of Covid-19. Stud Health Technol Inform 2021;281:880-884 [FREE Full text] [CrossRef]
    5. Kong Q, Riedewald D, Askari M. Factors affecting portal usage among chronically ill patients during the COVID-19 pandemic in the Netherlands: cross-sectional study. JMIR Hum Factors 2021 Jul 19;8(3):e26003 [FREE Full text] [CrossRef] [Medline]
    6. Osborn C, Mayberry L, Wallston K, Johnson K, Elasy TA. Understanding patient portal use: implications for medication management. J Med Internet Res 2013 Jul 03;15(7):e133 [FREE Full text] [CrossRef] [Medline]
    7. Irizarry T, DeVito Dabbs A, Curran CR. Patient portals and patient engagement: a state of the science review. J Med Internet Res 2015 Jun 23;17(6):e148 [FREE Full text] [CrossRef] [Medline]
    8. Ammenwerth E, Neyer S, Hörbst A, Mueller G, Siebert U, Schnell-Inderst P. Adult patient access to electronic health records. Cochrane Database Syst Rev 2107;6:CD012707 [FREE Full text] [CrossRef]
    9. Essén A, Scandurra I, Gerrits R, Humphrey G, Johansen M, Kierkegaard P, et al. Patient access to electronic health records: differences across ten countries. Health Policy Technol 2018 Mar;7(1):44-56 [FREE Full text] [CrossRef]
    10. Aanestad M, Grisot M, Hanseth O, Vassilakopoulou P. Strategies for building ehealth infrastructures. In: Aanestad M, Grisot M, Hanseth O, Vassilakopoulou P, editors. Information Infrastructures within European Health Care: Working with the Installed Base. Cham: Springer International; 2017:35-51.
    11. Walker J, Leveille S, Bell S, Chimowitz H, Dong Z, Elmore J, et al. OpenNotes after 7 years: patient experiences with ongoing access to their clinicians’ outpatient visit notes. J Med Internet Res 2019 May 06;21(5):e13876 [FREE Full text] [CrossRef]
    12. Zanaboni P, Kummervold P, Sørensen T, Johansen MA. Patient use and experience with online access to electronic health records in Norway: results from an online survey. J Med Internet Res 2020 Feb 07;22(2):e16144 [FREE Full text] [CrossRef] [Medline]
    13. Graham T, Ali S, Avdagovska M, Ballermann M. Effects of a web-based patient portal on patient satisfaction and missed appointment rates: survey study. J Med Internet Res 2020 May 19;22(5):e17955 [FREE Full text] [CrossRef] [Medline]
    14. de Lusignan S, Mold F, Sheikh A, Majeed A, Wyatt J, Quinn T, et al. Patients' online access to their electronic health records and linked online services: a systematic interpretative review. BMJ Open 2014 Sep 08;4(9):e006021 [FREE Full text] [CrossRef] [Medline]
    15. Mold F, de Lusignan S, Sheikh A, Majeed A, Wyatt J, Quinn T, et al. Patients’ online access to their electronic health records and linked online services: a systematic review in primary care. Br J Gen Pract 2015 Mar 02;65(632):e141-e151 [FREE Full text] [CrossRef]
    16. Otte-Trojel T, de Bont A, Rundall T, van de Klundert J. How outcomes are achieved through patient portals: a realist review. J Am Med Inform Assoc 2014;21(4):751-757 [FREE Full text] [CrossRef] [Medline]
    17. Tang P, Ash J, Bates D, Overhage J, Sands DZ. Personal health records: definitions, benefits, and strategies for overcoming barriers to adoption. J Am Med Inform Assoc 2006 Mar 01;13(2):121-126 [FREE Full text] [CrossRef]
    18. Wiljer D, Urowitz S, Apatu E, DeLenardo C, Eysenbach G, Harth T, Canadian Committee for Patient Accessible Health Records. Patient accessible electronic health records: exploring recommendations for successful implementation strategies. J Med Internet Res 2008 Oct 31;10(4):e34 [FREE Full text] [CrossRef] [Medline]
    19. Yamin C, Emani S, Williams D, Lipsitz S, Karson A, Wald J, et al. The digital divide in adoption and use of a personal health record. Arch Intern Med 2011 Mar 28;171(6):568-574 [FREE Full text] [CrossRef] [Medline]
    20. Delone W, McLean E. The DeLone and McLean model of information systems success: a ten-year update. J Manag Inf Syst 2014 Dec 23;19(4):9-30 [FREE Full text] [CrossRef]
    21. Ronda M, Dijkhorst-Oei LT, Rutten GEHM. Reasons and barriers for using a patient portal: survey among patients with diabetes mellitus. J Med Internet Res 2014 Nov 25;16(11):e263 [FREE Full text] [CrossRef] [Medline]
    22. Valeur H, Lie A, Moen K. Patient rationales against the use of patient-accessible electronic health records: qualitative study. J Med Internet Res 2021 May 28;23(5):e24090 [FREE Full text] [CrossRef] [Medline]
    23. Walker D, Hefner J, Fareed N, Huerta T, McAlearney AS. Exploring the digital divide: age and race disparities in use of an inpatient portal. Telemed J E Health 2020 May;26(5):603-613 [FREE Full text] [CrossRef] [Medline]
    24. Heponiemi T, Jormanainen V, Leemann L, Manderbacka K, Aalto AM, Hyppönen H. Digital divide in perceived benefits of online health care and social welfare services: national cross-sectional survey study. J Med Internet Res 2020 Jul 07;22(7):e17616 [FREE Full text] [CrossRef] [Medline]
    25. Powell KR. Patient-perceived facilitators of and barriers to electronic portal use: a systematic review. Comput Inform Nurs 2017;35:565-573 [FREE Full text] [CrossRef]
    26. van der Vaart R, Drossaert C, Taal E, Drossaers-Bakker K, Vonkeman H, van de Laar MAFJ. Impact of patient-accessible electronic medical records in rheumatology: use, satisfaction and effects on empowerment among patients. BMC Musculoskelet Disord 2014 Mar 26;15:102 [FREE Full text] [CrossRef] [Medline]
    27. Moll J, Rexhepi H, Cajander Å, Grünloh C, Huvila I, Hägglund M, et al. Patients' experiences of accessing their electronic health records: national patient survey in Sweden. J Med Internet Res 2018 Nov 01;20(11):e278 [FREE Full text] [CrossRef] [Medline]
    28. Turvey C, Klein D, Fix G, Hogan T, Woods S, Simon S, et al. Blue Button use by patients to access and share health record information using the Department of Veterans Affairs' online patient portal. J Am Med Inform Assoc 2014;21(4):657-663 [FREE Full text] [CrossRef] [Medline]
    29. Sääskilahti M, Aarnio E, Lämsä E, Ahonen R, Timonen J. Use and non-use of a nationwide patient portal: a survey among pharmacy customers. J Pharm Health Serv Res 2020;11:335-342 [FREE Full text] [CrossRef]
    30. Woods S, Forsberg C, Schwartz E, Nazi K, Hibbard J, Houston T, et al. The association of patient factors, digital access, and online behavior on sustained patient portal use: a prospective cohort of enrolled users. J Med Internet Res 2017 Oct 17;19(10):e345 [FREE Full text] [CrossRef] [Medline]
    31. Helsper EJ. A corresponding fields model for the links between social and digital exclusion. Commun Theor 2012 Oct 15;22(4):403-426 [FREE Full text] [CrossRef]
    32. van Deursen AJ, van Dijk JA. The first-level digital divide shifts from inequalities in physical access to inequalities in material access. New Media Soc 2019 Feb;21(2):354-375 [FREE Full text] [CrossRef] [Medline]
    33. Graetz I, Gordon N, Fung V, Hamity C, Reed ME. The digital divide and patient portals: internet access explained differences in patient portal use for secure messaging by age, race, and income. Med Care 2016;54:772-779 [FREE Full text] [CrossRef]
    34. Heponiemi T, Gluschkoff K, Leemann L, Manderbacka K, Aalto A, Hyppönen H. Digital inequality in Finland: access, skills and attitudes as social impact mediators. New Media Soc 2021 Jul 28:146144482110230 [FREE Full text] [CrossRef]
    35. European Commission. DESI: Shaping Europe’s Digital Future.   URL: https://digital-strategy.ec.europa.eu/en/policies/desi [accessed 2021-12-20]
    36. Jormanainen V, Parhiala K, Niemi A, Erhola M, Keskimäki I, Kaila M. Half of the Finnish population accessed their own data: comprehensive access to personal health information online is a corner-stone of digital revolution in Finnish health and social care. FinJeHeW 2019 Nov 02;11(4):289-310 [FREE Full text] [CrossRef]
    37. Kyytsönen M, Aalto AM, Vehko T. Social and Health Care Online Service Use in 2020–2021: Experiences of the Population (English abstract; report 7/2021). Finland: Finnish Institute for Health and Welfare; 2021.
    38. Jormanainen V, Reponen J. CAF and CAMM analyses on the first 10 years of national Kanta services in Finland. FinJeHeW 2020 Dec 23;12(4):302-315 [FREE Full text] [CrossRef]
    39. Kanta. Data Recorded in Kanta.   URL: https://www.kanta.fi/en/data-in-kanta [accessed 2022-04-19]
    40. Kanta. My Kanta Pages.   URL: https://www.kanta.fi/en/my-kanta-pages [accessed 2022-04-19]
    41. Aarnio E, Huupponen R, Martikainen J, Korhonen MJ. First insight to the Finnish nationwide electronic prescription database as a data source for pharmacoepidemiology research. Res Social Adm Pharm 2020 Apr;16(4):553-559 [FREE Full text] [CrossRef] [Medline]
    42. Jormanainen V. Large-scale implementation and adoption of the Finnish national Kanta services in 2010–2017: a prospective, longitudinal, indicator-based study. FinJeHeW 2018 Dec 04;10(4):381-395 [FREE Full text] [CrossRef]
    43. Kanta. Review of Kanta Year 2020: Importance of Digital Services Highlighted.   URL: https:/​/www.​kanta.fi/​en/​web/​guest/​notice/​-/​asset_publisher/​cf6QCnduV1x6/​content/​katsaus-kanta-vuoteen-2020-digipalvelujen-tarkeys-korostui [accessed 2022-04-19]
    44. Kanta. Statistics.   URL: https://www.kanta.fi/en/statistics [accessed 2022-04-19]
    45. Kanta. Acting on Behalf of Someone Else.   URL: https://www.kanta.fi/en/acting-on-behalf-of-someone-else [accessed 2022-04-19]
    46. Finnish Institute for Health and Welfare. National FinSote Survey.   URL: https:/​/thl.​fi/​en/​web/​thlfi-en/​research-and-development/​research-and-projects/​national-finsote-survey [accessed 2021-11-08]
    47. Seaman S, White IR. Review of inverse probability weighting for dealing with missing data. Stat Methods Med Res 2013 Jun;22(3):278-295 [FREE Full text] [CrossRef] [Medline]
    48. Härkänen T, Kaikkonen R, Virtala E, Koskinen S. Inverse probability weighting and doubly robust methods in correcting the effects of non-response in the reimbursed medication and self-reported turnout estimates in the ATH survey. BMC Public Health 2014 Nov 06;14:1150 [FREE Full text] [CrossRef] [Medline]
    49. Scholaro. Finland Grading System.   URL: https://www.scholaro.com/pro/Countries/Finland/Grading-System [accessed 2021-11-26]
    50. Marsh C, Elliott J. Exploring Data: An Introduction to Data Analysis for Social Scientists, 2nd Edition. Hoboken, NJ: Wiley; 208.
    51. Statistics Finland. Statistical Grouping of Municipalities.   URL: https://www.stat.fi/meta/kas/til_kuntaryhmit_en.html [accessed 2021-09-30]
    52. Miilunpalo S, Vuori I, Oja P, Pasanen M, Urponen H. Self-rated health status as a health measure: the predictive value of self-reported health status on the use of physician services and on mortality in the working-age population. J Clin Epidemiol 1997 May;50(5):517-528 [FREE Full text] [CrossRef]
    53. Jyväsjärvi S. Frequent Attenders in Primary Health Care. A Cross-Sectional Study of Frequent Attenders’ Psychosocial and Family Factors, Chronic Diseases and Reasons for Encounter in a Finnish Health Centre. University of Oulu.   URL: http://jultika.oulu.fi/files/isbn9514264460.pdf [accessed 2022-04-19]
    54. van Deursen AJ, Helsper E, Eynon R. Development and validation of the Internet Skills Scale (ISS). Inf Commun Soc 2015 Aug 25;19(6):804-823 [FREE Full text] [CrossRef]
    55. Ranganathan P, Pramesh C, Aggarwal R. Common pitfalls in statistical analysis: logistic regression. Perspect Clin Res 2017;8(3):148-151 [FREE Full text] [CrossRef] [Medline]
    56. Ammenwerth E, Duftschmid G, Al-Hamdan Z, Bawadi H, Cheung N, Cho K, et al. International comparison of six basic eHealth indicators across 14 countries: an eHealth benchmarking study. Methods Inf Med 2020 Dec;59(S 02):e46-e63 [FREE Full text] [CrossRef] [Medline]
    57. Lämsä E, Timonen J, Mäntyselkä P, Ahonen R. Pharmacy customers' experiences with the national online service for viewing electronic prescriptions in Finland. Int J Med Inform 2017 Jan;97:221-228 [FREE Full text] [CrossRef] [Medline]
    58. Sääskilahti M, Ahonen R, Timonen J. Pharmacy customers' experiences of use, usability, and satisfaction of a nationwide patient portal: survey study. J Med Internet Res 2021 Jul 16;23(7):e25368 [FREE Full text] [CrossRef] [Medline]
    59. Sääskilahti M, Ojanen A, Ahonen R, Timonen J. Benefits, problems, and potential improvements in a nationwide patient portal: cross-sectional survey of pharmacy customers' experiences. J Med Internet Res 2021 Nov 03;23(11):e31483 [FREE Full text] [CrossRef] [Medline]
    60. Hargittai E. Digital na(t)ives? Variation in internet skills and uses among members of the “Net Generation”. Sociol Inq 2010;80:92-113 [FREE Full text] [CrossRef]
    61. Marco-Ruiz L, Bønes E, de la Asunción E, Gabarron E, Aviles-Solis J, Lee E, et al. Combining multivariate statistics and the think-aloud protocol to assess human-computer interaction barriers in symptom checkers. J Biomed Inform 2017 Oct;74:104-122 [FREE Full text] [CrossRef] [Medline]
    62. Dopelt K, Avni N, Haimov-Sadikov Y, Golan I, Davidovitch N. Telemedicine and eHealth literacy in the era of COVID-19: a cross-sectional study in a peripheral clinic in Israel. Int J Environ Res Public Health 2021 Sep 10;18(18):9556 [FREE Full text] [CrossRef] [Medline]
    63. Kontos E, Blake K, Chou WYS, Prestin A. Predictors of eHealth usage: insights on the digital divide from the Health Information National Trends Survey 2012. J Med Internet Res 2014 Jul 16;16(7):e172 [FREE Full text] [CrossRef] [Medline]
    64. Hoogenbosch B, Postma J, de Man-van Ginkel JM, Tiemessen N, van Delden JJM, van Os-Medendorp H. Use and the users of a patient portal: cross-sectional study. J Med Internet Res 2018 Sep 17;20(9):e262 [FREE Full text] [CrossRef] [Medline]
    65. Hörhammer I, Kujala S, Hilama P, Heponiemi T. Building primary health care personnel's support for a patient portal while alleviating eHealth-related stress: survey study. J Med Internet Res 2021 Sep 22;23(9):e28976 [FREE Full text] [CrossRef] [Medline]
    66. Kerns J, Krist A, Longo D, Kuzel A, Woolf SH. How patients want to engage with their personal health record: a qualitative study. BMJ Open 2013 Jul 30;3(7):e002931 [FREE Full text] [CrossRef] [Medline]
    67. Krist A, Woolf S, Bello G, Sabo R, Longo D, Kashiri P, et al. Engaging primary care patients to use a patient-centered personal health record. Ann Fam Med 2014;12(5):418-426 [FREE Full text] [CrossRef] [Medline]
    68. Kujala S, Rajalahti E, Heponiemi T, Hilama P. Health professionals’ expanding ehealth competences for supporting patients’ self-management. Stud Health Technol Inform 2018;247:181-185.
    69. Zhao J, Song B, Anand E, Schwartz D, Panesar M, Jackson G, et al. Barriers, facilitators, and solutions to optimal patient portal and personal health record use: a systematic review of the literature. AMIA Annu Symp Proc 2018;2017:1913-1922.
    70. Clarke M, Lyden E, Ma J, King K, Siahpush M(, Michaud T, et al. Sociodemographic differences and factors affecting patient portal utilization. J Racial Ethn Health Disparities 2021 Aug;8(4):879-891 [FREE Full text] [CrossRef] [Medline]
    71. Tieu L, Schillinger D, Sarkar U, Hoskote M, Hahn K, Ratanawongsa N, et al. Online patient websites for electronic health record access among vulnerable populations: portals to nowhere? J Am Med Inform Assoc 2017 Apr 01;24(e1):e47-e54 [FREE Full text] [CrossRef] [Medline]
    72. Eriksson-Backa K, Hirvonen N, Enwald H, Huvila I. Enablers for and barriers to using My Kanta: a focus group study of older adults' perceptions of the National Electronic Health Record in Finland. Inform Health Soc Care 2021 Dec 02;46(4):399-411 [FREE Full text] [CrossRef] [Medline]
    73. Kruse C, Bolton K, Freriks G. The effect of patient portals on quality outcomes and its implications to meaningful use: a systematic review. J Med Internet Res 2015 Feb 10;17(2):e44 [FREE Full text] [CrossRef] [Medline]
    74. Pyper C, Amery J, Watson M, Crook C. Patients’ experiences when accessing their on-line electronic patient records in primary care. Br J Gen Pract 2004;54(498):38-43.
    75. Finlex. Laki digitaalisten palvelujen tarjoamisesta 306/2019.   URL: https://www.finlex.fi/fi/laki/alkup/2019/20190306 [accessed 2022-03-31]
    76. W3C Web Accessibility Initiative. W3C Accessibility Standards Overview.   URL: https://www.w3.org/WAI/standards-guidelines/ [accessed 2022-03-31]
    77. Alajarmeh N. Evaluating the accessibility of public health websites: an exploratory cross-country study. Univers Access Inf Soc 2021 Jan 27:1-19 [FREE Full text] [CrossRef] [Medline]
    78. Kanta. Accessibility Statement for the kanta.fi Web Service.   URL: https://www.kanta.fi/en/web/guest/accessibility-statement-for-the-kanta.fi-web-service [accessed 2022-04-19]
    79. Kanta. Get to Know Your My Kanta Pages.   URL: https://www.kanta.fi/en/my-kanta-pages-online-school [accessed 2022-01-19]
    80. Wildenbos G, Peute L, Jaspers M. Facilitators and barriers of electronic health record patient portal adoption by older adults: a literature study. Stud Health Technol Inform 2017;235:308-312.
    81. Czaja S, Zarcadoolas C, Vaughon W, Lee C, Rockoff M, Levy J. The usability of electronic personal health record systems for an underserved adult population. Hum Factors 2015 May;57(3):491-506 [FREE Full text] [CrossRef] [Medline]
    82. Kahn JS, Aulakh V, Bosworth A. What it takes: characteristics of the ideal personal health record. Health Aff (Millwood) 2009;28(2):369-376 [FREE Full text] [CrossRef] [Medline]
    83. Ministry of Social Affairs and Health. Information to Support Well-Being and Service Renewal. eHealth and eSocial Strategy 2020.   URL: https://julkaisut.valtioneuvosto.fi/handle/10024/74459 [accessed 2022-01-18]
    84. Luque A, van Keken A, Winters P, Keefer M, Sanders M, Fiscella K. Barriers and facilitators of online patient portals to personal health records among persons living with HIV: formative research. JMIR Res Protoc 2013 Jan 22;2(1):e8 [FREE Full text] [CrossRef] [Medline]
    85. Bidmead E, Marshall A. A case study of stakeholder perceptions of patient held records: the Patients Know Best (PKB) solution. Digit Health 2016;2:2055207616668431 [FREE Full text] [CrossRef] [Medline]
    86. Masur PK. How online privacy literacy supports self-data protection and self-determination in the age of information. Media Commun 2020 Jun 23;8(2):258-269 [FREE Full text] [CrossRef]

    ‎
    IPW: inverse probability weighting
    OR: odds ratio

    Edited by C Lovis; submitted 24.02.22; peer-reviewed by T Risling, J Haverinen; comments to author 27.03.22; revised version received 01.04.22; accepted 11.04.22; published 22.04.22

    Copyright

    ©Emma Kainiemi, Tuulikki Vehko, Maiju Kyytsönen, Iiris Hörhammer, Sari Kujala, Vesa Jormanainen, Tarja Heponiemi. Originally published in JMIR Medical Informatics (https://medinform.jmir.org), 22.04.2022.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Medical Informatics, is properly cited. The complete bibliographic information, a link to the original publication on https://medinform.jmir.org/, as well as this copyright and license information must be included.