The PRISMA flow diagram for the selected studies in the search process and the eligibility assessment are summarized in (Fig. 1). The initial electronic database search led to 1542 potentially relevant citations in the form of a title, abstract, bibliography, and full-text research. After removal of duplicates and initial screening, 125 articles were selected for further evaluation via full-text articles. Of these full-text articles, 103 articles were excluded due to the following reasons: 38 studies reported the prevention of SARS other than COVID-19; 36 have measured prevention measures other than contact tracing, screening, quarantine, and isolation; 19 had inappropriate study designs (commentaries, letters and case reports); and 10 were reviews or protocols. Thus, 22 studies [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35] met the inclusion criteria and were included in the systematic review.
The 22 studies [14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35] that were retained for the final analysis were published in the period from January 15, 2020, to June 02, 2020, based on participant populations in the following countries: China (n = 10), UK (n = 4), USA (n = 2), Hong Kong (n = 2), and Netherlands, Japan, France, and Taiwan (n = 1 from each). The included studies comprised of 9 observational [14,15,16,17,18,19,20,21,22] and 13 modeling studies [23,24,25,26,27,28,29,30,31,32,33,34,35]. With duplicates (repeated count), 3 of the studies assessed the overall prevention strategies [21,22,23], 5 assessed the effect of contact tracing [14, 24, 25, 33, 35], 2 assessed screening strategies [17, 34], 12 assessed the effect of quarantine [15, 23,24,25,26,27,28,29,30,31], and 6 assessed the effect of isolation [17, 25, 26, 31, 33, 35]. The sample sizes in the studies varied from hundreds to millions. Four studies were investigated for effect at the health facility level, while the remaining 18 studies explored at the community or national level. Survey characteristics and summary results are described in Table 1.
Quality (risk of bias) assessment within included studies
Summaries of the risk of bias assessment of non-randomized studies and quality rating of the modeling studies are presented in Tables 2 and 3, respectively. Two studies [14, 19] have low bias due to confounding, eight studies have low bias in selection of participants into the study, and all studies have low bias in classification of interventions. The overall risk of bias is moderate for eight studies and serious for one study. On the other hand, we have no concern for nine modeling studies, and two studies have major concerns.
COVID-19 prevention strategies and effectiveness
The summary result is presented in Table 1. Among the nine observational studies, three of them assessed COVID-19 transmission with the existing prevention measures at a community level in Taiwan, China, and Hong Kong [18,19,20]. The other two studies assessed the effect of escalating prevention measures at health facilities in China and Hong Kong [21, 22], and three studies [15,16,17] assessed national- and metropolitan-based quarantine strategies and the effect of laboratory-based quarantine in the prevention of COVID-19. The last study evaluated the effect of community-based contact tracing in UK .
The three studies [18,19,20] that assessed the overall prevention strategies found out that integration of interventions need to be applied instead of adhering to a single intervention. Cheng  reported that isolating symptomatic patients alone may not be sufficient enough to contain the epidemic. Wang  and Law  also concluded that in intimate contacts the transmission is 40–60%. Preventing contact through different strategies and integration is very important.
Studies conducted on the effect of quarantine [15,16,17] found that it can have a massive preventive effect. One of the studies  that assessed the effect of quarantine in different populations and quarantine strategies found that it should be integrated with input population reduction (travel restriction), and the other study  that assessed the effects of metropolitan-wide quarantine on the Spread of COVID-19 in China found that quarantine would prevent 79.27% (75.10–83.45%) of deaths and 87.08% (84.68–89.49%) of infections. Also, the other researcher  evidenced that laboratory-based screenings accomplished within hours can enhance the efficiency of quarantine.
Two studies described infection control preparedness measures in health care settings of Hong Kong and China [21, 22]. One of these studies  reported that infection transmission is highly increased within a short period of time and multiplicity of infection prevention strategies were recommended for prevention in health care setups. The other study  also concluded that practicing working shift among professionals working in facilities can be used as strategy to prevent thetransmission of COVID infection.
A study conducted by Keeling et al.  assessed the efficacy of contact tracing for the containment of COVID-19 in the UK. The study evaluated the contact pattern of the community and concluded that rapid contact tracing to reduce the basic reproduction number (R0) from 3.11 to 0.21 enables the outbreak to be contained. Additionally, it was found that each new case requires an average of 36 individuals to be traced, with 8.7% of cases having more than 100 close traceable contacts.
In this review, we identified 13 modeling studies [23,24,25,26,27,28,29,30,31,32,33,34,35] that assessed the effectiveness of contact tracing, screening, quarantine, and isolation for prevention of COVID-19 in different settings and groups. The simulation was done in individual or group basis and with different assumptions. Most of these studies used a model parameter from Chinese reports.
Three of these researches [25,26,27] particularly emphasized on the way how the R0 can be reduced and the epidemic would be reduced. The simulation by Tang et al.  aimed to estimate the R0 of SARS-CoV-2 and infer the required effectiveness of isolation and quarantine to contain the outbreak. Their susceptible-exposed-infected-recovered (SEIR) model estimated R0 of 6.47 and generalized that 50% reduction of contact rate achieved by isolation and quarantine would decrease the confirmed cases by 44%; reducing contacts by 90% also can decrease the number of cases by 65%. The other researcher, Rocklov (27), by using data from the Diamond Princess Cruise ship, concluded that quarantine of passengers prevented 67% of cases and lowered the R0 from 14.8 to 1.78. Similarly, the reduction of R0 was achieved from quarantine .
In addition to these, five studies [24, 28, 30, 31, 35] which modeled the effectiveness of different public interventions consistently reported that integrated intervention is better than a single intervention. One of these research conducted in the UK  found that combined isolation and tracing strategies would reduce transmission more than mass testing or self-isolation alone (50–60% compared to 2–30%). The other study  also reported that with R0 of 2.4, a combination of case isolation and voluntary quarantine for 3 months could prevent 31% of deaths. The others also concluded that quarantine should be strict and integrated with contact tracing, screening, and other interventions [30, 31, 35].
Five modeling studies also assessed the effect of quarantine [23, 29, 32], contact tracing , and screening . All of the studies [23, 29, 32] reported that quarantine has reduced the incidence of infection and shortened the duration of the epidemic. However, the effectiveness depends on the level of integration with other strategies. Similarly, model simulations that assessed the effect of contact tracing and screening reported that the strategies are effective. However, as the report of Hellewell  stated, contact tracing and isolation might not contain outbreaks of COVID-19 unless very high levels of contact tracing are achieved. Similarly, the other researcher  reported that in a stable epidemic, under the assumption that 25% of cases are subclinical, it is estimated that arrival screening alone would detect roughly one-third of infected travelers.