Phone-Delivered Lifestyle Coaching Helped Older Adults Stay Healthier During COVID-19

When the first wave of the COVID-19 pandemic swept across Europe in March 2020, older adults faced the sharpest end of the threat: higher rates of hospitalization, intensive care admission, and death driven by an age-related decline in immune function and the chronic conditions that accumulate with age. Public health agencies urged people to stay active, eat well, quit smoking, and moderate their drinking, but almost no randomized evidence existed to show whether structured lifestyle counselling could actually change behavior in this vulnerable group, let alone protect them from severe outcomes. A nationwide Hungarian trial now offers some of the clearest answers to date, and its findings are both encouraging and sobering.

The trial, known as PROACTIVE-19, was launched at the height of the first pandemic wave with an ambitious design. Researchers planned to enroll 3,800 participants per arm and randomly assign adults aged 60 and older to receive either general health education or personalized lifestyle counselling delivered entirely by telephone. The intervention targeted five domains: mental health, smoking habits, physical activity, dietary habits including body mass index assessment, and alcohol consumption. General counselling followed standard public health guidelines in roughly 10-minute calls, while personalized counselling tailored recommendations to each individual’s lifestyle patterns in 15 to 20 minute sessions. Follow-up assessments were conducted weekly during the first month, biweekly during the second month, and monthly thereafter, allowing researchers to track behavioral trajectories with unusual granularity for a trial of this scale.

Delivering the intervention required remarkable legislative flexibility. Hungarian law was amended through a special government decree to allow participants to provide verbal consent by phone during the very first call, accommodating the restrictions imposed by the pandemic. Enrollment began on March 26, 2020, across Hungary, with participants recruited through television, radio, and social media advertisements directed to a toll-free phone number. Calls were delivered by trained personnel including medical doctors, dietitians, psychologists, and research administrators, with call quality supervised by a Steering Committee. In total, 3,975 calls were performed across the study period, creating one of the most detailed longitudinal behavioral datasets assembled during the pandemic.

Ultimately, 441 participants were analyzed, with 214 assigned to the general counselling arm and 227 to the personalized arm. The study was terminated early after reaching only about 5 percent of the planned sample size, following a predefined stopping rule triggered by the successful introduction of SARS-CoV-2 vaccines. This early closure had significant consequences for the trial’s statistical power. Only one confirmed SARS-CoV-2 infection was recorded during intervention follow-up, making it impossible to compare the prespecified COVID-19 severity endpoint, a composite of intensive care admission, hospitalization lasting more than 48 hours, and mortality among participants with incident infection. Instead, the researchers reported all-cause mortality obtained from national records, with follow-up extending to March 17, 2023.

The survival analysis revealed no statistically significant difference between the two arms. Median follow-up for all-cause mortality was 930 days in the general arm and 937 days in the personalized arm. A Cox proportional hazards model estimated a hazard ratio of 0.66 favoring the personalized group, with a 95 percent confidence interval of 0.28 to 1.53, but the result did not reach statistical significance. The researchers noted that statistical power for the survival analysis was only 16 percent, severely limiting the interpretation of null findings. The proportional hazards assumption was verified using Schoenfeld residuals, confirming no evidence of violation, but the low event count meant that multivariable adjustment was not performed.

Where the trial truly shines is in its longitudinal behavioral analysis. Using linear mixed-effects models with random intercepts to account for within-subject correlation, and quantifying parameter uncertainty with percentile wild bootstrap confidence intervals based on 2,000 resamples, the researchers identified several significant improvements over time. Fruit and vegetable consumption increased significantly across the full sample, with a time effect of 0.052 units per call. Smoking amount decreased significantly, particularly among participants who were current smokers at baseline, where the time effect was −0.372 cigarettes per day. Psychological variables also showed meaningful change, though the direction varied by domain.

The personalized intervention showed one clear advantage over general counselling: a significantly greater reduction in estimated alcohol consumption among baseline drinkers, with a time-by-group interaction of −1.357. This effect remained significant in sex-stratified analyses for women but not for men. However, the personalization came with an unexpected drawback. Among participants with obesity, body mass index increased in the personalized group, with a more pronounced interaction effect of 0.068 compared to the full sample. The researchers suggest that the tailored dietary counselling may have inadvertently encouraged increased food intake in some individuals, though the exact mechanism remains unclear.

Physical activity showed no significant change over time in the full sample, and the time-by-group interaction was not statistically significant. Graphical trends visually favored the personalized arm in both men and women, but these differences were not supported by the model results. Seasonal and pandemic-wave effects substantially influenced physical activity and psychological distress. Participants joining in autumn exhibited significantly more negative slopes in both physical activity and fruit and vegetable consumption compared with those joining in spring. COVID-related fear was significantly lower in summer compared with spring, while physical activity levels were significantly higher in autumn. Compared with the first pandemic wave, physical activity was significantly higher outside wave one, and COVID-related fear was significantly lower, indicating that the initial wave was associated with reduced activity levels and higher psychological distress.

The trial’s findings carry important implications for pandemic preparedness and healthy ageing. While the additional benefit of personalization was limited overall, the significant improvements observed across several lifestyle domains, including diet, smoking, and alcohol consumption, emphasize the importance of maintaining and promoting healthy behaviors during pandemic situations. The researchers note that unhealthy lifestyle factors in combination accounted for 51 percent of severe COVID-19 in the UK population, and that adherence to healthy lifestyle prior to infection has been shown to reduce the risk of post-COVID-19 condition. The trial also demonstrates the real-world feasibility of telemedicine-delivered lifestyle interventions at a system level, requiring legislative adaptation but ultimately reaching a geographically diverse population.

Several limitations warrant consideration. The early termination reduced statistical power, particularly for infrequent clinical outcomes and subgroup analyses. Several outcomes were self-reported, and COVID-19-specific mortality and potentially mild infections could not be reliably captured. The open-label nature precluded participant blinding, though participants from the same household were not enrolled to minimize contamination between groups. No correction for multiple comparisons was applied across outcomes, and seasonality and pandemic-wave effects were not included in the primary models, although secondary analyses suggested these contextual factors influenced several results. Despite these constraints, the trial provides randomized evidence supporting the potential role of lifestyle-based, non-pharmacological strategies in maintaining overall health during pandemic conditions, and highlights the importance of contextual factors when interpreting behavioral changes in older adults.

Subject of Research: Effect of telemedicine-delivered lifestyle interventions in older adults during the COVID-19 pandemic: a nationwide two-arm randomized trial (PROACTIVE-19)

Article Title: Effect of telemedicine-delivered lifestyle interventions in older adults during the COVID-19 pandemic: a nationwide two-arm randomized trial (PROACTIVE-19)

Article References: Váncsa, S., Nagy-Szakolczai, A., Molontay, R., Miklós, E., Szentesi, A., Ocskay, K., Hegyi, P. J., Hegyi, J., Hegyi, A., Hegyi, E., Czapári, D., Molnár, Z., Erőss, B., Hegyi, P., PROACTIVE-19 Trial Contributors, Abonyi-Tóth, Z., Birkás, B., Csathó, Á., Czopf, L., … Zsigmond, B. (2026). Effect of telemedicine-delivered lifestyle interventions in older adults during the COVID-19 pandemic: a nationwide two-arm randomized trial (PROACTIVE-19). GeroScience. https://doi.org/10.1007/s11357-026-02585-1

Image Credits: AI Generated

DOI: 10.1007/s11357-026-02585-1

Keywords: PROACTIVE-19, telemedicine, lifestyle intervention, older adults, COVID-19, randomized controlled trial, GeroScience, healthy ageing, Hungary, behavioural change, Effect, telemedicine-delivered