Baseline characteristics
A total of 2,128 individuals from Israel responded to the survey between July 2021 and August 2022. Table 1 provides a description of demographic factors, comorbidities, habits, and socioeconomic variables available for the study participants. The participants had a mean age of 49.83 years, with 61.14% being female and 75.94% identifying as Jewish, which is in line with the national average. Among them, 823 individuals (38.67%) tested positive for SARS-CoV-2. Of the infected participants, 744 (90.4%) reported symptoms, whereas 79 (9.6%) were asymptomatic. Based on the reported infection dates of the participants, the peak of COVID-19 infections occurred in January 2022, coinciding with the fifth wave driven by the Omicron variant.
Regarding BMI distribution, 23.2% of participants were classified as obese (BMI > 30, Table 1). At the time of being surveyed, 43.4% had received three or more doses of a COVID-19 vaccine and 22.6% were not vaccinated. In terms of chronic conditions, 12.6% reported hypertension, followed by 7.5% with diabetes. Other chronic conditions such as asthma (4.7%), COPD (2%), chronic kidney disease (2.3%), and immunosuppression (2.7%) were also reported (Table 1).
Comparison with published literature
Based on our search, 408 articles were initially identified from the electronic database searches. After scanning titles and abstracts, 361 articles were excluded for lack of relevance. This left 47 potentially relevant MAs for full review. Of these, 27 MAs were excluded due to lack of relevance and failure to adhere to PRISMA guidelines. Consequently, 20 MAs were retained (see Fig. 1S). Of these, 7 MAs did not match our prespecified criteria for outcomes, while 13 MAs did: 3 MAs matched infection outcomes, 5 matched hospitalization outcomes and 5 matched both infection and hospitalization outcomes. Of these 13 MAs, six examined risk factors that were obtained in our dataset and were therefore selected for direct comparison with our findings.
The impact of epidemiological factors on COVID-19 infection rates
In the univariate analysis, gender, age, number of children, ethnicity, education, BMI, vaccination status, hypertension, COPD, smoking, alcohol consumption and physical activity were significantly associated with COVID-19 susceptibility (p < 0.05, Table 2).
In the multivariate logistic regression model, after adjusting for other factors, ethnicity, education level, BMI, vaccination status and the presence of COPD remained significantly associated with COVID-19 susceptibility (Table 2). BMI and vaccination status remained significantly associated with COVID-19 infection in the subsequent time period (supplementary Table 1).
Our study did not find a significant association between gender and susceptibility to infection. A published MA of 41 studies [9] reported that men have a slightly higher risk of infection RR = 1.14 (95% CI: 1.07–1.21) (Figs. 2S-A). Similarly to our study, 14 of the 41 papers included did not identify an association.
Regarding differences between minorities and the general population, the Arab minority was more likely to be infected than the Jewish majority (aOR = 1.48, 95% CI: 1.02–2.15), a result similar to other settings where minority groups experience higher susceptibility. A MA of 68 studies showed that African American and Hispanic individuals were almost two times as likely to test positive for COVID-19 as White individuals (African American: RR, 2.01; 95% CI, 1.04–3.88; P = 0.04; Hispanic: RR, 2.09; 95% CI, 1.13–3.88; P = 0.02) [11].
In our study, participants with a BMI over 30 had increased odds of COVID-19 infection compared to those with a normal BMI (aOR = 1.54, 95% CI: 1.02–2.32), a finding compatible with a MA of 7 studies [12] reported an OR of 2.73 (95% CI: 1.53–4.87) for COVID-19 infection among individuals with higher BMI (Fig. 2S-B). Our study also found that being underweight was associated with increased susceptibility (Table 2) but we could not identify a MA specifically focusing on low BMI as a risk factor.
Our results suggested a significantly reduced likelihood of infection among vaccinated individuals, with 2 doses (aOR = 0.09, 95% CI: 0.06–0.13, p < 0.0001), a protective effect compatible with a pooled estimate of a MA of 21 studies [2] (Fig. 2S-C), reporting a RR of 0.19 (95% CI: 0.13–0.27, p < 0.0001).
A MA of 17 studies [18] suggested a reduced risk of COVID-19 infection among smokers, with a RR of 0.74 (95% CI of 0.58–0.93) (Fig. 2S-D). Although our OR was very similar (0.82), smoking was not associated with a change in the odds of COVID-19 infection in our study, like 8/17 studies included in the MA.
A MA of 5 studies [17] reported a small but statistically significant reduction in infection risk associated with physical activity (RR = 0.89, 95% CI: 0.84–0.95) (Fig. 2S-E). Our study did not identify any effect (OR = 0.69, p > 0.05).
To assess the consistency of our findings across different phases of the pandemic, we conducted a supplementary analysis comparing associations between key variables and infection status before and after January 1st 2022 (supplementary Table 1). While the primary analysis was restricted to infections occurring before this date, when testing was more systematic and the force of infection lower, we recognize that risk factor dynamics may evolve over time. The analysis showed that while certain associations (e.g., BMI, physical activity, vaccination status) were consistent across both periods, overall the total number of significant associations decreased after January 1st 2022, possibly reflecting the difficulty in identifying risk factors in a high force of infection scenario where almost everyone is infected. Findings of this additional analysis should be interpreted with caution.
The impact of epidemiological factors on COVID-19 severity (hospital admission rates)
In the univariate analysis, gender, age, ethnicity, education, BMI and vaccination status were significantly associated with COVID-19 severity as measured by hospitalization (p < 0.05, Table 3). After adjusting for significant confounders, age, education level, and vaccination status remained significantly associated with the severity of the disease.
A MA of 22 studies [9](Figure 3S-A) identified an elevated risk of hospitalization among males (OR =1.33, p<0.05). Although our study had a similar OR (1.48), the association was not significant. Likewise, a MA of 5 studies [12](Figure 3S-B) reporting an overall positive association between obesity and disease severity (OR=1.72, CI: 1.55-1.92, p<0.05). Our study found a similar OR of 1.82 for the association between obesity (BMI>30) and COVID-19 hospitalization, though the association was not significant.
Our study found a decrease of 21% in the odds of hospitalization in vaccinated individuals (2 doses and 3 or more doses) and a 63% decrease for those who received 3 or more doses, compared to unvaccinated individuals, compared to unvaccinated (95% CI:0.41-1.52 and CI: 0.17-0.81, respectively), compatible with the findings of a MA of 7 studies [26] (Figure 3S-C) showing a 93% reduction with full vaccination (2 doses; RR =0.07 (95% CI: 0.03-0.17).