Background
Existing evidence suggests the high burden of the COVID-19 pandemic on patients with non-communicable diseases (NCDs). Individuals with NCDs have a higher risk of COVID-19, but pandemic control policies have resulted in significant disruptions to the screening, treatment, and surveillance of NCD patients [
1‐
3]. In May 2020, 155 of the WHO Member States participated in a rapid assessment of the impact of the COVID-19 pandemic on NCDs resources and services. Approximately ¾ of countries reported a considerable degree of disruption to NCD services [
4]. The need to maintain continuity of care for patients with NCDs to ensure the optimum level of disease control and management has been recognized as a priority during the COVID-19 pandemic, requiring adjustments to daily routines and the development of new care models.
In Latvia, 14-day COVID-19 cases and death notification rates were among the lowest in the EU countries during the first half of 2020 [
5]. However, upon assessment of the epidemiological risks in neighbouring and other EU countries, a state of emergency due to the first wave of the COVID-19 pandemic throughout the country was declared from 12 March until 9 June 2020 [
6]. Along with other regulations, on 27 March, it was decided to restrict planned inpatient health care services and secondary outpatient health care services, with only lifesaving health care services and those that required continuity of treatment being offered. Disruptions of planned health care services raised concerns about chronic patients’ access to care. Premature mortality from chronic conditions such as malignant neoplasms, diabetes mellitus, ischaemic heart disease, and chronic lower respiratory disease is one of the highest in Latvia among the EU member states [
7].
The potential for remote consultations and digital tools to improve access to health services has been recognized before [
8‐
10]. However, technical, administrative, financial, and professional barriers have limited the uptake and use of remote consultations [
11,
12]. The COVID-19 pandemic has prompted the use of innovative ways to continue outpatient care for patients with NCDs, resulting in a sharp increase in the use of remote counselling [
13]. Telemedicine has been acknowledged as one of the solutions to ensure health monitoring and disease management for patients with NCDs and to protect vulnerable populations with a high risk of infection and negative health outcomes [
14].
Before the state of emergency, in the pre-crisis phase, remote counselling was not routinely provided and covered by public health funding in Latvia. The decisions to reimburse GPs and specialists for remote consultations were passed on the 18th and 27th of March 2020, respectively [
15,
16]. The use of formal remote consultations for patient care was a new experience for clinicians in Latvia’s primary and secondary health care practice. This study was designed to evaluate the changes in the structure and volume of healthcare consultations for patients with selected NCDs by paying particular attention to the roles and conditions of remote health services.
This study aimed to describe the dynamics of consultations and the volume of remote consultations provided for patients with particular NCDs and to explore clinicians’ experiences of providing remote consultations during the first wave of the COVID-19 pandemic in Latvia.
Methods
The study was carried out as part of the VPP-COVID-2020/1–0011 project of the National Research programme to assess the impact of COVID-19 on health care and public health in Latvia.
Study design
This study utilized a mixed-method approach using a convergent design as described by Creswell and Clark [
17]. The study analysed quantitative data from the National Health Services (NHS) and employed inductive thematic analysis of qualitative data from semi-structured interviews with clinicians. Both data sets were collected and analysed in parallel. Integration through narrative was used at the interpretation and reporting level; the findings were analysed separately, using a contiguous approach, and merged in the discussion [
17,
18].
Quantitative study
Data from the Latvia NHS were used to quantitatively illustrate the change in health service provision. All residents of Latvia are entitled to state-funded health care, and the NHS is responsible for contracts with health care providers and payments for publicly funded services. The use of private health care is not recorded in the NHS data and, therefore, this variable was not included in the analysis. When any service is reported to the NHS, the main diagnosis is indicated according to the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10). The recording of comorbidities is optional; there is no identifier used for suspected or confirmed cases of the disease. To characterize the care of NCDs, the following conditions (ICD-10 codes) were selected: 1) breast cancer (C50) in women and prostate cancer (C61) in men; 2) type 1 and type 2 diabetes (E10 and E11); 3) hypertension, coronary heart disease and congestive heart failure (I10, I20-I25, and I50); and 4) chronic obstructive pulmonary disease (COPD) and asthma (J44 and J45). All records of patient consultations in the entire adult population with these codes were extracted from the NHS Database of Outpatient Services alongside with the speciality of the medical professional who provided the service. Records from GPs were analysed for all abovementioned conditions, whereas specialists were selected for each group of diseases as follows: surgeons for breast cancer, urologists for prostate cancer, endocrinologists for diabetes, cardiologists for hypertension, coronary heart disease, and heart failure, and pulmonologists and/or allergists for chronic obstructive pulmonary disease and asthma.
The main focus was on the number of consultations provided in the first half of 2020; nevertheless, the same period in 2019 was also included to assess the expected pattern of service use. The first weeks were excluded because of their different lengths, and the New Year holidays as well as the last weeks in June were excluded because they were public holidays. Thus, three main periods were defined: Period I, before the emergency (the weeks of January 6 to March 15 in 2020 were compared with the weeks of January 7 to March 17 in 2019); Period II, when restrictions were implemented (the weeks of March 16 to May 10 in 2020 were compared with the weeks of March 18 to May 12 in 2019); and Period III, when restrictions, including consultations with specialists, were gradually lifted (the weeks of May 11 to June 21 in 2020 were compared with May 13 to June 23 in 2019). Period II also covered the Eastern Holidays, as well as public holidays on the 1st and 4th of May in both years. The adult population of Latvia on the 1st of January in 2019 and 2020 was used to calculate the rate of service use. Because of the different lengths of the studied periods, the consultation rate was calculated as the number of consultations in the given period per 100,000 persons per day (population size multiplied by the number of days in the given period). The proportion of remote consultations was calculated as the number of remote consultations in the period divided by the total number of consultations in that period. OpenEpi software [
19] was used to estimate 95% confidence intervals (CIs) for rates (Mid-P exact test), rate differences (Byar’s method) and proportions (Wilson Score Interval).
Qualitative study
Participants
Purposive maximum variation sampling was employed across clinicians to ensure the spread and heterogeneity of the clinicians’ views.
A purposive maximum variation sampling ensures heterogeneity of research participants with a wide range of variation across different key criteria (age, gender, experience, region, discipline areas etc.) [
20] to ensure the maximum variability within the data relevant to the phenomenon under examination [
21].
Qualitative studies aim to identify the qualitatively different patterns observed in data rather than to quantify those patterns [
22]. The question of sample size in a qualitative study can be difficult, as there are no precise guidelines in this area [
23] and a broad range of suggestions of the numbers of participants needed for qualitative interviews are provided [
22]. Generally, a sample size of between 15 and 30 individual interviews is mentioned as adequate in qualitative research striving to identify patterns across primary data-set [
24]. The concept of ‘saturation’ is an appropriate tool for determining sample size in qualitative research [
24].
Participants were selected according to the following sampling criteria: age; gender; place of practice (city or suburb/rural area); level of health care (primary or secondary); clinician type (GP or specialist physician); type of health care facility (ambulatory (outpatient care centre), hospital or both); and clinician work experience (in years). Table
1 presents the demographic data and characteristics of the study participants.
Table 1
Characteristics of the interviewed participants (n = 34)
Level of health care and speciality |
Primary care: general practitioner | 19 | 55.88 |
Secondary care: specialist physician | 15 | 44.12 |
cardiologist | 3 | 20.00 |
oncologist | 3 | 20.00 |
internist | 3 | 20.00 |
endocrinologist | 3 | 20.00 |
pulmonologist | 3 | 20.00 |
Type of health care facility (institution) |
only ambulatory | 22 | 64.70 |
only hospital | 6 | 17.65 |
both ambulatory and hospital | 6 | 17.65 |
Place of practice |
republican cities | 22 | 64.70 |
other | 12 | 35.30 |
Clinician work experience (in years) |
< 10 | 4 | 11.76 |
10–20 | 11 | 32.35 |
21–40 | 13 | 38.24 |
40< | 6 | 17.65 |
Age |
25–45 | 12 | 35.30 |
46–65 | 16 | 47.05 |
66< | 6 | 17.65 |
Gender |
women | 28 | 82.35 |
men | 6 | 17.65 |
Participants were recruited via professional associations for healthcare workers and health care institutions. Recruitment began with the development of an official informative invitation letter for clinicians to participate in the study. The letter was then distributed to the relevant professional associations and health care institutions with a request to inform clinicians about the study. A total of thirty-four clinicians were interested in participating in the study and agreed to the interview.
Data collection
The semi-structured interview method was used for qualitative data collection in the present research. Before the interviews, an interview guide was developed and piloted. The interview guide included questions on the clinicians’ experiences of adapting practice management to the COVID-19 context; experiences of providing care for NCDs patients in an emergency due to COVID-19; perceptions of patients’ health needs; attitudes and beliefs about remote consulting; and perceptions of the benefits and risk factors associated with the care of NCD patients in a state of emergency, including the risks and benefits of remote consultation.
Data were collected between August 2020 and November 2020.
Six researchers who had no prior relationship with the research participants participated in the interview process.
The interviews were conducted individually; i.e., one researcher and one research participant met either in person or remotely via video communication or phone depending on the interviewee’s preference and the epidemiological situation in the country. Of the 34 interviews conducted, 20 were face-to-face interviews, and 14 were remote interviews (via phone, Skype, or Zoom). All face-to-face interviews were undertaken at the different health care centres where the research participants worked. No one except the participant being interviewed and the interviewer was present at the interviews. One semi-structured interview was conducted with each study participant. The researchers did not observe the need for repeated interviews.
Before each interview, written consent was provided, and the researcher repeated the purpose of the study. Participation in the study was completely voluntary. The participants were informed that they could withdraw from the interview at any time for any reason.
The interviews lasted between 35 and 90 min and were audio-recorded with participant permission.
Data analysis
The interviews were transcribed verbatim in preparation for the analysis.
To explore a broad scope of participant experiences, inductive thematic analysis was performed with the transcriptions, and the data were analysed manually using a six-step method as described by Braun and Clarke [
25]: 1) familiarisation with data, 2) generating initial codes, 3) searching for emerging themes, 4) reviewing the themes, 5) defining and naming the themes, and 6) writing up the analysis and producing the report. Each interview transcript was coded independently by two researchers of the research team (MK and IS1). The team members discussed emergent themes, agreed on a final version of the list of general themes and subsequent sub-themes, and interpreted them.
Inductive thematic analysis was selected because this method encourages the development of themes exclusively based on the data as opposed to the adjustment of themes to a pre-selected thematic framework or predetermined theory [
25,
26].
Discussion
The COVID-19 pandemic abruptly changed the context in which the health care system operated and posed new challenges for the continuity of health care provision for patients with NCDs [
27,
28]. The emergency during the first wave of the COVID-19 pandemic in Latvia affected not only the number and structure of consultations for patients with selected NCDs but also new health care counselling practices among health care providers. A significant decrease in health care consultations by both GPs and specialists for selected NCDs occurred when the restrictions on planned health care services were most severe (the weeks of March 16 to May 10, 2020). The provision of consultations for cancer-related conditions was less affected; however, counselling for breast and prostate cancer by both GPs and specialists declined, even though cancer care was exempted from the health service restrictions. The qualitative findings indicate that patients’ fear of contracting COVID-19 was one of the factors contributing to their decision to delay consulting with a doctor even as their symptoms worsened or new symptoms manifested.
Remote consultations, which accounted for a significant share of the total number of consultations during the first wave of the COVID-19 pandemic, were essentially a new type of health care service in Latvia. There are countries where telemedicine services have been around for decades, such as the United Kingdom where primary care providers have been encouraged to develop new, flexible models of patient access, including remote types of consultation in the 2022 GP Action Plan [
29]. In Latvia, reporting on remote consultations to the NHS was subserved by reimbursement, which was introduced only in the second half of March 2020. In spring 2020, 31.6–51.1% of GPs consultations for particular NCDs in Latvia were provided remotely. Other studies reported various results with remote consultations making 34–89% of all consultations provided by GPs, with a higher proportion reached in health care systems with pre-established and routinely-used telemedicine [
30,
31]. Some degree of underreporting of remote consultations in Latvia is very likely, particularly among GPs, because of the work intensity. The qualitative findings highlight increased time resources required to register remote consultations and eHealth functionality issues, which could hinder the reporting.
Despite the initiation of remote consultations, the total volume of consultations provided by both GPs and specialists was significantly lower for most of the selected NCDs compared with the same period in 2019, experiencing up to a 15% decrease in GP consultations for prostate cancer and diabetes and a 43% decrease in pulmonologist consultations. Moreover, when the restrictions on health services were lifted, the proportion of remote consultations declined, with a more significant decrease observed for specialist consultations (Table
4). Several emerging themes in qualitative findings illustrate obstacles for remote consultations to be seen as a routine form of patient consultation. One of the commonly mentioned complexities of remote consultations was the ability to make the right treatment decision without face-to-face contact and physical examination of a patient. Similar concerns, especially regarding patients with a new set of symptoms, were expressed in other studies [
32]. Fewer concerns were raised regarding well-known patients. The findings in this study are consistent with previously published data indicating some reluctance among GPs to implement remote consultations due to concerns about medico-legal aspects and technical support; however, if these issues can be addressed, clinicians are open to the development of telemedicine in clinically appropriate situations [
33‐
35]. The necessity for a well-established framework for remote consultations with easy-to-use administrative systems, clear guidelines and centralized technological support was noted by the clinicians in the interviews. These findings align with conclusions from other studies that indicate the necessity of a strategic approach and the involvement of all main stakeholders in the development of telemedicine services [
36,
37]. Remote consultations must be considered a separate health care service, not an on-site consultation performed remotely, and must be planned and organized accordingly.
In particular, a necessity for enhancements in the eHealth system emerged as a separate theme in our findings with special significance regarding remote consultations. The development of an eHealth system in Latvia started in 2005, but the use of the central eHealth system is still voluntary for medical institutions. Since January 2018, electronic sick-leave certificates and prescriptions for state reimbursed pharmaceuticals have been mandatory. Health care institutions are obliged to provide online information on referrals for inpatient or hospital services, radiological examinations, outpatient examinations, and vaccinations [
38]. Unfortunately, today, the eHealth system is affected by a broad variety of technical and technological problems, as well as legal and data security issues, resulting in the low commitment of healthcare organizers and providers to use the system. Detailed analysis of the problems faced by the eHealth platform has already been published elsewhere [
39].
The COVID-19 pandemic has challenged health care systems globally [
27,
40]. It has also shed light on the myriad of long-term health care problems in Latvia, e.g., the adversities in the development and operability of the eHealth system, the lack of clearly regulated telemedicine services and the low health and digital literacy of patients and healthcare providers, as revealed by our study findings described above. The need for the mentioned issues to be resolved has been more visible, as these issues play a critical role in ensuring timely access to appropriate healthcare for the Latvian population, not only in times of crisis but also beyond, especially in the remote parts of the country. The Latvian healthcare system and policymakers have recognized these realities and considered the lessons of this crisis by developing a basis for legislation on the provision, registration and compensation of remote consultations [
41] and announcing the development of a new, modern, up-to-date eHealth platform [
42]. Many issues remain to be addressed to fully solve the identified problems; nonetheless, by prompting the establishment of telemedicine services as a permanent and well-functioning part of the Latvian healthcare system in the future and by simultaneously demanding physical distancing and increased access to healthcare services, similar to other countries [
43,
44], the COVID-19 crisis has unintentionally served as a powerful catalyst for fundamental long-awaited improvements in Latvian healthcare. However, in the future, the performance and effectiveness of implemented eHealth services, tools, and measures should be evaluated in detail.
Limitations
This study has several limitations that should be taken into account. The analysis of data on provided health care consultations from the NHS database covers the whole population, but consultations that occurred in the private sector were not included. Data on the volume and dynamics of consultations provided in the private sector were not available and should be viewed as a potential confounder. Regarding services in the public sector and their reporting to the NHS, the validity of diagnostic codes is not systematically evaluated. In addition, these codes may be used for any tests or referrals related to suspected, not confirmed, conditions. Moreover, our analysis did not include consultations provided to patients with chronic diseases for reasons other than the selected disease codes. For example, NCD patients might have consulted GPs because of acute respiratory symptoms or any other health issues that were coded using other disease codes, and these consultations would therefore have not been analysed. Furthermore, GPs decision on the main diagnosis to be reported in routine consultations with multimorbid patients could be a confounder. Consequently, there is a possibility that consultation volume distribution in-between diagnoses analysed was affected since the recording of comorbidities is optional. However, the NHS database provides access to country-level data on health care services allowing the assessment of the impact of the COVID-19 emergency on the national health care system. Regarding the qualitative data, maximum variation sampling criteria did not include specialists and GPs working in a private sector without a contract with the NHS, as well as specialists representing other fields of medicine (dermatology, rheumatology, geriatrics etc.) therefore their perspective was not explored. All GPs and specialists who participated in the study were volunteers, and thus possibly more extroverted and willing to express their experiences, and more research-oriented. Qualitative interviews were performed a few months after the end of the first wave of the COVID-19 emergency. Therefore, the accuracy of the clinicians’ reports of their experiences might have been affected by recall bias. Nonetheless, the qualitative analysis enhanced our in-depth understanding of clinicians’ experiences of consultations in the context of the COVID-19 crisis and restrictions of health care services in Latvia.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.