Table 1. Baseline model parameters, with distributions and sources.
Parameter | Distribution | Mean (with standard deviation) or range or confidence interval (CI) | Sources |
---|---|---|---|
Humans | |||
Population at risk in base year | Triangular | Country-specific |
Most likely value is based on [33]; minimum value based on [34]; maximum value is based on [35] |
Crude birth rate of population (annual) | Deterministic | Country-specific | [36] |
Growth rate of urban population (annual) | Deterministic | Country-specific | [37] |
Crude death rate (annual) | Deterministic | Country-specific | [38] |
Probability of developing severe dengue after secondary infection (weekly) | Beta | 0.048 (0.013) | [39], consistent also with [12], but we conservatively assume a zero probability after primary infection |
Probability of death among severe dengue patients, without case management (weekly) | Uniform | Min = 0.05 or country-specific Max = 0.20 |
[40][41], with floor set at maximum of: 0.05 or country-specific probability of death among severe dengue patients, with case management |
Probability of death among severe dengue patients, with case management (weekly) | Beta | Country-specific | [32] |
Average age of patients who die from severe dengue | Deterministic | 30 | Conservative assumption, since the majority of deaths tend to occur among children, aged <15 years.[21] |
Average life expectancy of general population | Deterministic | Country-specific | [42] |
Rate of loss of maternal antibodies (weekly) | Gamma | 0.055 (0.041) | [12] |
Average duration of infectiousness, primary and secondary (weeks) | Deterministic | 1 | [12]; duration of illness differs (see below). |
Average duration of recovery from primary infection, including cross immunity (years) | Lognormal | 1.88 (95% CI 0.88–4.31) | [43], gives a weekly probability of 0.011, conservative relative to [12] which used rate of 0.0055 per day or a weekly probability of 0.038 |
Relative probability of being susceptible to a second infection | Triangular | Min = 0.1 Most likely = 0.66, Max = 0.75 |
Maximum is comparable to [12] which put the relative probability of being susceptible to ith infection at (5-i)/4 |
Dengue disability weight | Beta | 0.197 (0.004) | [44], conservative relative to [9] and [45], and same as [12] |
Severe dengue disability weight | Beta | 0.545 (0.004) | [44], conservative relative to [9] and [45], same as [12]; we assumed a correlation of 0.999 with non-severe dengue so that dengue disability weight does not exceed the severe dengue disability weight |
Average duration of dengue illness (years) | Beta | 0.019 (0.007) | [12] |
Average duration of severe dengue illness (years) | Beta | 0.034 (0.009) | [12]; assumed correlation of 0.999 with non-severe dengue |
Discount rate for health | Uniform | 0–3% | [46] |
Vectors | |||
Average number of female vectors per host | Uniform | 1.0–6.0 | [27], with assumed sine function of period length 2π and a standard deviation of 20% |
Increase in vector populations during outbreak | Uniform | 100–200% | [29] |
Outbreak duration (weeks), no outbreak control | Uniform | 2.0–9.0 | Assumption |
Outbreak periodicity (years) | Triangle | Min = 2 Most likely = 3 Max = 5 |
Conservative assumption relative to [30] |
Death rate of adult vector (weekly) | Lognormal | 0.255 (0.010) | [47], which is between the values used by [12] for young and old adults |
Biting rate (bites per day) | Lognormal | 0.561 (0.129) | [48], conservative relative to [12], which used 0.7; conservative assumption of correlation of -0.5 with vectors per host |
Probability of transmission: vector to human (per bite by an infected vector) | Beta | 0.572 (0.083) | [48], conservative relative to [12], which used 0.9 |
Probability of transmission: human to vector (per bite of an infected human) | Beta | 0.493 (0.067) | [48] |
Resource use and costs | |||
Proportion of cases hospitalized, non-severe dengue (primary hospital) | Beta | 0.14 (0.03) | [49] |
Proportion of cases hospitalized, severe dengue (secondary hospital) | Deterministic | 1.00 | [50] |
Duration of hospital stay, non-severe dengue (bed days) | Gamma | 3.84 (0.64) | [5][51] |
Ratio of duration of hospital stay, severe dengue to non-severe dengue | Deterministic | 1.5 | [52] |
Ambulatory visits for a hospitalized case (number) | Gamma | 4.42 (0.81) | [53] [51] [54] |
Ambulatory visits for a non-hospitalized case (number) | Gamma | 3.68 (0.76) | [51] [53] [54], with assumed correlation of 0.999 with hospitalized cases |
Unit cost of hospital bed day, primary hospital (2013 US$) | Lognormal | Country-specific | See Table 3 |
Unit cost of hospital bed day, specialist hospital (2013 US $) | Lognormal | Country-specific | See Table 3; assumed correlation of 0.999 with primary hospital |
Unit cost of ambulatory clinic visit (2013 US$) | Lognormal | Country-specific | See Table 3 |
Discount rate for costs | Uniform | 3–6% | [46] |