Table 1.
Input Parameters for Base Case Simulation and Distributions of Parameters
| Input Parameter | Base Case Value | Sensitivity/Uncertainty Analysis | Source |
|---|---|---|---|
| Infection prevalence in the hospital | 0.02 | Triangular distribution (min = 0, max = 0.5, mode = 0.2) | National Records Scotland,7 Public Health Scotland,22Scottish government31(estimated) |
| Infection prevalence in the community | 0.05 | Triangular distribution (min = 0, max = 0.2, mode = 0.05) | Perez-Reche and Strachan21 |
| The probability that an infected resident dies (age-specific) | Drawn for each individual resident from empirical distribution by age: 80+ y: 11% 70–79 y: 6.0% 60–69 y: 2.6% 50–59 y: 0.71% 40–49 y: 0.18% 30–49 y: 0.09% 20–29 y: 0.04% 18–20 y: 0.007% |
No (This parameter does not impact our main model output, the number of infected residents, significantly.) | Ferguson et al,13Kulu and Dorey32
The Infection Fatality Rate (IFR) for Scotland is adjusted based on the overall aged-adjusted IFR value for the UK and the relative IFR value (=1.18) for other urban areas in Scotland. Most of the population (>80%) in North Lanarkshire live in areas classified as other urban areas. |
| The probability that an infected staff member dies | Drawn for each individual staff member from a uniform distribution (0.0003–0.022) | No | Ferguson et al,13Kulu and Dorey32 |
| The no. of contacts that a resident has with other residents per day | Drawn for each individual resident from a Poisson distribution with a mean of 3.9 contacts per resident per day |
Mean of the Poisson distribution is drawn from a triangular distribution (min, 1; max, 5; mode = 3.9) | Van den Dool et al,11Chamchod and Ruan33 |
| The no. of contacts that a staff has with other staff per day | Drawn for each individual staff member from a Poisson distribution with a mean of 7.3 contacts per staff member per day | Mean of the Poisson distribution is drawn from a triangular distribution (min, 1; max, 10; mode, 7.3) | Van den Dool et al11 |
| The no. of contacts that a staff has with residents per day | Drawn for each individual staff member from a Poisson distribution with a mean of 16.2 contacts per staff per day | Mean of the Poisson distribution is drawn from a triangular distribution (min, 10; max, 20; mode, 16.2) | Van den Dool et al,11Chamchod and Ruan33 |
| The no. of contacts that a staff has with visitors per day | 5.0 contacts per staff member per day | Triangular distribution (min, 0; max, 10; mode, 5.0) | Discussions with the manager and staff of the representative care home |
| The probability that a resident comes into contact with another resident in the other unit | 20% | Triangular distribution (min, 0; max, 0.5; mode, 0.2) | Discussions with the manager and staff of the representative care home |
| The average no. of people visiting a resident per day | 1.0 visitor per resident per day | Triangular distribution (min, 0; max, 2.0; mode, 1.0) | Van den Dool et al,11Port et al34 |
| The rate at which residents leave the care home because of deaths caused by other reasons, moving to another facility, admitted to hospitals, or returning to their own home (rare) | 0.005 deaths or discharges per resident per day | Triangular distribution (min, 0.001; max, 0.005; mode, 0.004) | Scotland Information Services19 (Calculated from data for care homes in North Lanarkshire) |
| Staff turnover rate | 24% per year | Triangular distribution (min, 0.1; max, 0.5; mod, 0.24) | Scottish Care35 |
| The probability that an infected resident will develop symptoms | Drawn for each individual resident from empirical distribution: 80+ y: 0.9 70–79 y: 0.85 60–69 y: 0.8 50–59 y: 0.75 40–49 y: 0.7 30–49 y: 0.65 20–29 y: 0.6 18–20 y: 0.55 |
Triangular distribution (min, 0.5; max, 0.9; mode, 0.8) | Ferguson et al,13Verity et al36 |
| The probability that an infected staff member will develop symptoms | 0.7 | Triangular distribution (min, 0.5; max, 0.9; mode, 0.7) | Ferguson et al,13Verity et al36
(for a population like the United Kingdom or the United States) |
| The probability that a symptomatic resident has severe symptoms | Drawn for each individual resident from empirical distribution: 80+ y: 0.28 70–79 y: 0.25 60–69 y: 0.17 50–59 y: 0.11 40–49 y: 0.05 30–49 y: 0.03 20–29 y: 0.01 18–20 y: 0.001 |
No (This parameter does not affect no. of infections significantly given the assumptions that symptomatic individuals are isolated) | Ferguson et al,13Kulu and Dorey32
The proportion of symptomatic cases requiring hospitalizations for Scotland is adjusted based on the overall aged-adjusted value for the United Kingdom |
| The probability that a symptomatic staff member has severe symptoms | Drawn for each individual staff member from a uniform distribution (0.01–0.17) | No | Ferguson et al,13Kulu and Dorey32 |
| The probability that an individual (resident or staff) is infected after coming into contact with another infectious individual (resident, staff or visitor) | 0.02 | Triangular distribution (min, 0.001; max, 0.1; mode, 0.02) | Ferguson et al,13Wang et al,37Tang et al38 |
| The time elapsed between first exposure and becoming infectious | Lognormal (μ = 1.16, σ = 0.85) | No (This parameter does not significantly affect number of infections as exposed individuals are not infectious. Also, values for this parameter are relatively consistent across studies.) | Lauer et al,39McAloon et al,40Nishiura et al41 (log normal mean, 4.6; SD, 4.8) |
| The time elapsed between becoming infectious and onset of symptoms | Discrete uniform distribution (1,3) | No (Values for this parameter are consistent across studies.) | He et al,42 Gatto et al,43 Byrne et al44 |
| The time elapsed between onset of symptoms and recovery (or recovery time for those who remain asymptomatic) | Asymptomatic: log normal (μ = 2.049, σ = 0.246) Symptomatic: Mild: log normal (μ = 2.049, σ = 0.246) Severe: log normal (μ = 2.624, σ = 0.170) |
No (There is a strong consensus about the distribution of this parameter in literature.) | Kerr et al,17 Wölfel et al45 |
| The reduction of resident–resident and staff–staff interactions when social distancing is implemented | 0.75 | Triangular distribution (min, 0.2; max, 0.9; mode, 0.75) | Assumed (based on other models’ assumption13,46 and discussions with care home staff and managers) |
| The sensitivity of RT-PCR test | 0.7 | Triangular distribution (min, 0.6; max, 0.98; mode, 0.7) | Watson et al,47 Arevalo-Rodriguez et al48 |
| The lag between testing and test result | 1 day | No (implemented in scenario-based uncertainty analysis) | Discussion with representatives from Public Health Medicine (NHS Lanarkshire) and Lanarkshire Health and Social Care Partnership |
| Effectiveness of isolation of infected residents | 100% | 50%, 75%, and 100% | Assumed (based on other models’ assumptions13,46) |