key: cord-0950824-rs1favrh
authors: Tatsioni, Athina; Siountri, Iliana; Tsamoulis, Donatos; Vafeidou, Kyriaki
title: Clinical trials during pandemic in primary care: Low number and low validity after one-year experience
date: 2021-10-11
journal: The European journal of general practice
DOI: 10.1080/13814788.2021.1986279
sha: 55d01cbb0aceec72d1391b121e30c513ebc21953
doc_id: 950824
cord_uid: rs1favrh

nan

During the pandemic, primary care (PC) core components, including access, comprehensiveness and continuity of services as well as person-centred care were challenged in all countries [1] [2] . Numerous articles on COVID-19 and PC, including opinion papers, narrative reviews, and descriptive studies, appeared in the literature. However, well-designed trials remain crucial to reduce uncertainty in harm-benefit ratio for PC interventions and facilitate risk communication. Searching in PubMed (February 2021), we identified only 10 trials [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] , and 1 pilot study conducted in PC (Supplementary Material) [13] . Based on these published trials, we try to delineate certain methodological issues that pertain to the components of a clinical trial, i.e. design, population, intervention, comparator and outcome. Subsequently, we suggest potential approaches that may overcome these issues in future PC trials.

Clinical trials in PC generally lacked a randomised design, and a comparator group. Notably, there were only two trials [3, 4] , including a control group; one of them with a randomised design (RCT) [3] . Non-controlled trials might have been informative on the feasibility of interventions. However, without a comparison group, they could hardly reach a robust conclusion on the effectiveness and safety.

In addition, PC trials usually did not focus on high-risk populations [14] [15] [16] [17] [18] , including older patients and people with underlying health conditions. There was only one trial on nursing home residents [9] ; and three trials on participants with chronic diseases [3, 4, 11] . Moreover, published trials did not focus on under privileged populations as defined by social determinants of health. Aggressive measures taken in most countries to mitigate COVID-19 led to reduced access, disruptive health care and significant social and economic repercussions [19] [20] [21] , especially where patients struggled with the technology and had increasing mental health issues. Previous reports warned of increased risk of morbidity and mortality from deferred referrals and routine PC services [22] [23] [24] .

Another methodological issue raised is the heterogeneity of evaluated interventions, with the majority designed as complex interventions. Trials recruiting patients with suspected and confirmed COVID-19 evaluated diverse interventions, including diagnostic modalities [6] [7] [8] [10] [11] [12] [13] , medications, telehealth use and multidisciplinary care models without including a control group. Evaluation of multi-component interventions to deliver preventive care services for non-COVID-19 conditions was practically missing except for a non-controlled trial assessing drive-through transcutaneous bilirubin screening for newborns [5] . Previous studies documented a decline in preventive services, such as childhood immunisation coverage and cancer screening [25] [26] [27] . There was no trial on interventions to promote or sustain COVID-19 vaccination. However, relevant trials may still be in progress.

Finally, another issue that poses difficulties in the interpretation of results is the type of assessed outcomes. Trials on suspected and confirmed COVID-19 patients [6-8,10-13] measured a variety of short-term outcomes, including patient health outcomes, such as symptom control, recovery, hospitalisations and mortality as well as health services outcomes, such as monitoring, cost, time and appropriate referral. Feasibility outcomes, including home monitoring and programme changes after feedback, were also reported. Few trials on patients with chronic illness evaluated both patient health outcomes [3, 4, 11] , i.e. depression symptom control, patient satisfaction and health services outcomes, i.e. visit attendance, level of integration of services. The only trial on preventive care focussed on feasibility and health services utilisation outcomes [5] .

Several recommendations may help overcome the methodological issues described above. In the initial phase of pandemic, the design of high-quality RCTs might be a difficult task. Researchers may justifiably avoid comparisons with placebo or no treatment during a pandemic. Instead, they may design trials including a control group to facilitate comparisons of complex interventions with usual care, i.e. telehealth versus in-person care, comparisons of timing of implementation, comparisons of different sequencing of interventions and comparisons of different combinations of interventions [28] . Intentto-treat estimations provided by RCTs are particularly relevant for interventions where adherence could be problematic or where the post-randomisation experiences of compared groups could be very different [29] , i.e. when multi-component interventions introduce several associated secondary changes.

Supplemental data for this article can be accessed here. In addition, eligibility criteria should consider other determinants of health [30, 31] , especially social factors that may jeopardise health care equity and affect outcomes. Besides high-risk populations, future RCTs recruiting under-privileged populations could provide robust evidence on specific multi-component interventions that may reduce inequalities. They may either compare precise implementation of interventions in high-risk populations and settings versus populations with lower risk or use risk information to target the populations and settings where interventions are to be applied [28] . Both short and long-term consequences due to health care inequalities during the pandemic may become evident in the years to come. Therefore, PC should have efficient interventions in place to support resilient health care.

To address issues due to heterogeneity of complex interventions, large RCTs may come closer to real-life circumstances using a pragmatic design, especially when assessing major clinical outcomes such as hospitalisations and death. For example, well-designed RCTs with or without a telehealth component may indicate how to best provide care for patients with acute and chronic illnesses as well as for patients in need of preventive and palliative care. Collaborative and integrated health care models need to be evaluated beyond the level of proof-of-concept study. Other designs potentially suitable for large PC trials include cluster RCTs of parallel clusters or steppedwedge RCTs [32] . During the design for RCTs, several issues may appear, such as how to consider carry-over effects from sequential treatments and interactions. Methodological recommendations based on experience from other medical fields, including crossover, adaptive and factorial RCT designs may be considered in PC [28, 33] . RCTs may also assess the impact of diagnostic modalities beyond diagnostic outcomes, evaluating their impact on therapeutic choice, patient and societal outcomes [34] . In addition, they may assess the impact of newly developed and validated clinical prediction models compared to usual clinical assessment [35] .

Despite the variety of outcomes, to evaluate effectiveness and safety, efforts should be made to measure outcomes that matter. PC may follow the paradigm of other medical fields (e.g. the Core Outcome Measures in Effectiveness Trials initiative http://www.comet-initiative. org/) in developing core outcome measures relevant to the care provided. Long-term outcomes are equally crucial to short-term outcomes. All trials except for the RCT by Hickey et al. [3] supported that their results encouraged further adoption. Future RCTs need to confirm replication of these results before widespread use. Besides, morbidity and mortality trials may also assess the impact of pandemic on other diseases, health care utilisation, social indicators (e.g. violence), quality of life and selfcare attainment.

The very small number of published trials, most of them without a control group, indicates that robust evidence on efficient PC interventions is not available one year after the pandemic started. An updated search in September 2021 showed that a few additional RCTs evaluating mainly drug interventions for patients with COVID-19 in the community appeared in PubMed. One may argue that most studies, both RCTs and others, may have limited usefulness [36, 37] . However, COVID-19 may offer PC an opportunity to use experience from research in other fields to design trials that have a maximum chance of being useful and informative.

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Family medicine in times of 'COVID-19': a generalists' voice

After the first wave: what effects did the COVID-19 measures have on regular care and how can general practitioners respond to this?

Pragmatic randomized trial of a pre-visit intervention to improve the quality of telemedicine visits for vulnerable patients living with HIV

System-wide accelerated implementation of telemedicine in response to COVID-19: mixed methods evaluation

Drive-through transcutaneous bilirubin screening for neonatal jaundice: a safe and efficient system during the COVID-19 pandemic

Rapid design and implementation of an integrated patient self-triage and selfscheduling tool for COVID-19

Primary care population management for COVID-19 patients

Use and Cost-Effectiveness of a telehealth service at a centralized COVID-19 quarantine center in Taiwan: cohort study

Medicalization of nursing facilities as a real alternative to hospitalization during the COVID-19 pandemic

Usefulness of lung ultrasound examinations performed by primary care physicians in patients with suspected COVID-19

Psychiatric nurse practitioners as leaders in behavioral health integration

Treating COVID-19 with colchicine in community healthcare setting

Feasibility and efficacy of home monitoring for patients with suspected or confirmed COVID-19

Age-dependent effects in the transmission and control of COVID-19 epidemics

Assessing the age specificity of infection fatality rates for covid-19: systematic review, meta-analysis, and public policy implications

Presymptomatic SARS-CoV-2 infections and transmission in a skilled nursing facility

COVID-19: what has been learned and to be learned about the novel coronavirus disease

Preliminary estimates of the prevalence of selected underlying health conditions among patients with coronavirus disease 2019 -United States

The untold toll -the pandemic's effects on patients without covid-19

Chinese mental health burden during the COVID-19 pandemic

Depression, anxiety, and stress and sociodemographic correlates among general Indian public during COVID-19

Covid-19: "staggering number" of extra deaths in community is not explained by covid-19

Reduction in STsegment elevation cardiac catheterization laboratory activations in the United States during COVID-19 pandemic

Reduced rate of hospital admissions for ACS during covid-19 outbreak in Northern Italy

Early impact of the coronavirus disease (COVID-19) pandemic and physical distancing measures on routine childhood vaccinations in England

Decline in child vaccination coverage during the COVID-19 pandemic -Michigan care improvement registry

Association of cancer screening deficit in the United States with the COVID-19 pandemic

Preserving equipoise and performing randomised trials for COVID-19 social distancing interventions

Randomized controlled trials: often flawed, mostly useless, clearly indispensable: a commentary on Deaton and Cartwright

Applying an equity lens to interventions: using PROGRESS ensures consideration of socially stratifying factors to illuminate inequities in health

Health promotion, inequalities and young people's health: a systematic review of research

The stepped wedge cluster randomised trial: rationale, design, analysis, and reporting

Adaptive clinical trials: advantages and disadvantages of various adaptive design elements

The efficacy of diagnostic imaging

Update to living systematic review on prediction models for diagnosis and prognosis of covid-19

Why most clinical research is not useful

Meta-research: bird's eye views of primary care research