Perception towards electronic health records & uptake of digital health IDs among the urban residents in northern India: A mixed methods study

Material & Methods

This study was undertaken at the department of Community Medicine and School of Public Health, Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, India, after obtaining clearance from the Institutional Ethics Committee of PGIMER, Chandigarh. Verbal informed consent was taken from all participants before administering the questionnaire. The interviews were undertaken in person. Audio consent was taken considering the COVID-19 pandemic situation at that time since the investigators wanted to minimize physical contact in context to a written consent form. The consent was recorded using a voice recorder after the participants read and/or were informed about the study.

The present study was a concurrent mixed method investigation. A cross-sectional design was adopted for the quantitative component and in-depth interviews (IDIs) for the qualitative part were conducted. The study participants were the adults visiting the routine outpatient department (OPD) and the community-based screening camps conducted in the coverage area of a primary health centre (PHC; field practice of the department) in Chandigarh. Adults who did not have Chandigarh-based Aadhar numbers and those who had already completed the study during their previous encounters were excluded. A PHC from Chandigarh was chosen because the study was undertaken during the piloting phase of the ABDM (2020-21), in which the government of India was only implementing it in the Union Territories (UT). This specific PHC area was chosen as per the memorandum of understanding with the UT health authorities, where the department can undertake research, provide healthcare services and implement national health programmes. The study was conducted from July 2021 to January 2022. Assuming a digital health ID uptake of 50 per cent among the eligible community, with an absolute error of five per cent and 95 per cent confidence levels, we calculated a minimum sample size of 400. Adding 10 per cent for the non-response rate, a sample size of 440 was required. The sampling technique was consecutive sampling of consenting eligible adults. In order to improve the representativeness, we kept a limit of ten participants per day in the OPD and ten participants per outreach camp to be enrolled in the study. It was done to ensure people availing different healthcare services (curative and preventive) at specific clinics (e.g., NCD clinics, antenatal clinics, immunization clinics) and camps (community-based NCD screening camps) on different days during the data collection period were included. Purposive sampling of individuals who wilfully declined (10 participants) to join ABDM by refusing to create health IDs was done to conduct the in-depth interviews. The total number of interviews was decided based on the saturation of information (Principle of redundancy).

Results

Latent class analysis

Three sub-groups (latent classes- LC) were identified by the LCA after testing various combinations of variables within the study population based on the following variables: EHR awareness, ABDM awareness, willingness to make their EHR part of the national EHR system for self-use, for policy-making or planning, for research use, feel that EHR is secure, overall in favour of EHR and allow doctors to access their EHR. The probability of each variable belonging to LCs is depicted in Figure 1 as proportions. The latent class 1 group of participants who had the least probability of being aware of EHR and ABDM were not willing to share their health records in EHR for policy and research and were not in favour of EHR creation for India. Latent class 2 characteristics included those equally unaware of EHR and ABDM, similar to latent class 1 but who were willing to allow EHR for self-use, policy, and research, felt secure about EHR in India, overall in favour of EHR and willing to allow the doctors to access complete records from their EHR. Latent class 3 characteristics included being well aware of EHR and ABDM, and also willing to allow EHR for self-use, policy, and research, feeling secure about EHR in India, overall in favour of EHR and willing to allow the doctors to access their EHR. A total of 77 (18.4%), 229 (54.6%) and 113 (27%) were classified under latent classes 1, 2 and 3, respectively. Compared to latent class group 1, group 2 [Prevalence Ratio (PR): 3.15; 95% Confidence interval (CI): 1.8-5.51] and group 3 (PR: 4.9; 95% CI: 2.42-9.93) had significantly higher uptake of digital health ID, while no significant difference between LC 2 and 3 was noted. Thus, LCA denotes that a favourable attitude towards the applications and security of the EHR ecosystem may lead to better uptake of digital health IDs and, subsequently, participation in the national EHR.

Close

Fig. 1.

Margin plot of the three latent class created in the study. EHR, electronic health records; NDHM, National Digital Health Mission.

Export to PPT

Barriers to digital health ID uptake

The qualitative analysis of in-depth interviews revealed a lack of awareness, apprehension towards the digital technology and perceived limited utility as the major themes for barriers to the uptake of digital health IDs (Fig. 2). A few verbatim of the non-acceptors of the digital health ID under the above themes are provided in the Box.

Close

Fig. 2.

Thematic analysis of the ‘barriers to uptake of ABDM-Health ID’. ABHA, Ayushman Bharat Health Account; IEC, Information, Education and Communication.

Export to PPT

Discussion

The digital health ID uptake rate was 78 per cent among the participants from Chandigarh. As of December 2023, 50.39 crores of ABHAs have been created across India.¹⁷ Chandigarh reported an ABHA coverage rate of 55 per cent. Among the larger States, Andhra Pradesh reported coverage of 77.6 per cent, the highest in the country¹⁷. The differential finding may be due to the period of this study. It was conducted during the pilot phase of the ABDM; thus, administrative focus and impetus were there on this specific programme, which might have plateau over time. The higher proportion of digital health ID uptake (80.4%) among the elderly population (>60 yr) underlines the role played by healthcare workers in assisting and enabling the elderly to create digital health IDs. The role of families and young individuals in enabling older individuals to access and avail the digital health services has also been explored to be useful in previous studies^18,19. Hence, it would be prudent to include the family members of the elderly within the target population for creating awareness about digital health IDs and EHRs.

Awareness regarding the EHR was low among the participants approaching the PHC in this study which was a significant determinant of digital health ID uptake. Furthermore, the media was found to be the major source of awareness for the participants who had heard about EHR. Qualitatively, awareness (rather a lack of it) about EHR was identified as a major barrier to digital health ID uptake, with refusers quoting the lack of communication through mass media. The same was observed in univariable analysis and the LC. Social media was seen as a reliable source by certain refusers as they had the option to read both the pros and cons from social media. These findings suggest that awareness of the EHR is essential, and multiple channels of communication should be used.

The security of health records in the EHR system was a major concern expressed in this study. A study from the UK²⁰ reported that 47 per cent of the people felt that EHR is less secure than the conventional method of health records storage. The confidence in the government’s ability to keep the EHR secure significantly impacted the digital health ID uptake in this study. Reduced confidence in government entities to ensure the safety of EHR data has also been expressed in UK²⁰. In the USA, over a period of time, participants showed a decreasing trend in privacy and security concerns regarding EHR usage²¹.

Apprehensions towards technology usage, trust in technology, inexperience, lack of digital literacy and privacy concerns were found to be barriers to adopting EHRs. Digital health literacy has been reported to be low in India (20%)²², and there is a lack of studies on this topic. Perceived lack of utility for digital health IDs and EHRs by the participants is another barrier, as it does not provide any incentives or motivation. The government of India is introducing an incentive scheme for healthcare institutes to create and link health records digitally under the ABDM from 2023 to improve EHR usage²³. In a study conducted on EMR in Chandigarh, participants reported a positive attitude towards the digitized settings as it had reduced their time in giving the previous history²⁴. Avoidance of repetition of examination and history taking was also reported as a potential positive factor for EHR acceptance in the UK²⁰.

The present study participants were willing to give doctors and pharmacists complete access to their health records, with identifiers, similar to the findings of a previous study from Korea²⁵. Most of our study participants were accommodative to share their health records with the researchers, while the majority refused to give access to any records to the drug companies. Similarly, research-related EHR access was accepted by a majority in the UK (81.38%). Refusing access to drug companies could be due to its potential for breach of privacy and associated commercial exploitation²⁰.

In our study settings, 74.7 per cent of the participants were in favour of setting up a national EHR, while in the UK, a slightly lower proportion was supportive of such an initiative (62.47 and 55%, respectively)^15,20. Allowing individuals to access their own digitalized health records has been reported as a potential benefit of the EHR²⁶, as well as improving patient-centred care²⁷. This could be explored in India’s digital health architecture.

The present study is the first to assess the patient and public perception towards EHRs in India in a programmatic setting. The mixed methods design has helped to identify and triangulate the major barriers to health ID uptake and EHR reluctance. Though the cross-sectional study design limits the temporality of association between the potential determinants identified, creation of the digital ID is a one-time event, and all participants who had accepted and created the digital health ID did so during the time of the data collection after enrolling in this study. Hence, the data which we collected under the variable regarding awareness of EHR was for the period prior to the creation of the digital health IDs. Hence, this may not be considered a limitation in the context of the current study. However, this limitation is applicable to EHR practices carried out by the participants. Limited generalizability of the findings to other Indian States, since the study was undertaken among people from a single PHC area, is another limitation. This was because the study was undertaken during the piloting phase of the ABDM (2020-21), in which the Government of India was implementing it in the UTs only. There is also potential selection bias since it was not a household-based study following probability sampling but rather undertaken majorly among those adults approaching an urban PHC. Although the authors modified and pre-tested the NHS-EHR tool in the current settings, it was not validated in India, which is another limitation. We applied LCA to identify the hidden population group characteristics that enabled the digital health ID uptake. However, the following limitations of the LCA apply to the current study as well: (a) reification, which indicates that some or all the identified sub-groups of the population by the LCA may not always exist in actual settings^28,29; (b)class assignment is not infallible³⁰; and (c) unknown impact of the study sample size on the latent classes²⁹. ‘Naming fallacy’ can also occur in LCA when the classes are specifically named, while it does not accurately describe the classes³⁰.

In conclusion, EHR awareness is a major determinant for the uptake of digital health IDs among urban residents of Chandigarh. Overall, a favourable attitude towards the applications and security of the EHR ecosystem among the public will lead to better uptake of digital health IDs and may increase participation in the national EHR. Campaigns in mass media and social media, as well as legal provisions addressing the above concerns, need to be conducted. Furthermore multi-centric studies involving various States of India must be undertaken to improve the generalizability of the current findings and to identify further context-specific determinants in those States. Nationally representative, serial cross-sectional surveys must be conducted to understand the pattern and trends of perception among the people as the EHR is rolled out and expanded in Indian settings. Digital health literacy and incentivization of EHR for its adoption need to be studied in the Indian context, which may impact the EHR ecosystem in the long run.