Table 1 Model parameters. The central estimates shown are derived from the source literature with ranges added for the sensitivity analysis.

From: Evaluating the impact of pulse oximetry on childhood pneumonia mortality in resource-poor settings

Parameter

Value (range)

Sources

Disease progression

Incidence

Country-specific ± 10%

Ref. 12

Proportion severe on day 1

5% (2–10%)

Ref. 13

Mean duration of non-severe illness before recovery

3 days (2–4 days)

Ref. 25

Mean duration of non-severe illness before progression to severe illness

10 days (9–11 days)

Estimated from model (see Methods)

Mean duration of severe illness before recovery

4 days (3–5 days)

Ref. 22

Mean duration of severe illness before death

7 days (6–8 days)

Ref. 26

Proportion bacterial versus viral (NSV)

85% viral (75–90%)

15% bacterial (25–10%)

Ref. 27

Proportion bacterial versus viral (SV)

85% bacterial (75–90%)

15% viral (25–10%)

Assumed

Care-seeking and health-care parameters

Mean duration of illness before care seeking

NSV 3 (2–4) days

SV 0.75 (0.5–1) days

Ref. 26

Probability that community-based treatment is available

Country-specific ± 10%

Ref. 3

Probability that timely hospital access

0.61 ± 10%

Ref. 24

Probability of community-based treatment curing non-severe bacterial case

0.925 (0.90–0.95)

Ref. 28

Probability of treatment with hospital care curing case

0.925 (0.80–0.95)

Assumed to be high if oxygen is available with lower values representing poorer standard of care

Probability of treatment with amoxicillin curing severe case if prescription adhered to

0.65 (0.6–0.7)

Ref. 29 (based on treatment failure rates of patients with hypoxia at baseline)

Prognostic parameters

Probability of prognostic available

1 (0.9–1)

Assumed to be high for the purpose of this analysis

Sensitivity of IMCI

0.55 (0.5–0.6)

Ref. 30

Sensitivity of PO1

0.7 (0.65–0.75)

Estimated

Sensitivity of PO2

0.85 (0.8–0.9)

Ref. 14

Specificity of IMCI

0.85 (0.8–0.9)

Assumed to be high given low overall referral rates

Specificity of PO1 and PO2

0.85 (0.8–0.9)

Assumed to be similar to IMCI

Adherence to non-severe prognosis (IMCI)

0.55 (0.5–0.6)

Refs 31,32,33

Adherence to severe prognosis (IMCI)

0.65 (0.6–0.7)

Refs 31,32

Adherence to non-severe prognosis (PO1 and PO2)

0.55 (0.5–0.6)

Assumed to be similar to IMCI

Adherence to severe prognosis (PO1 and PO2)

0.85 (0.8–0.9)

Assumed to be high for the purpose of this analysis

Prognosed SV treated with community-based treatment versus nothing

1

Assumed that prognosed SV will always be treated even if not referred to hospital

Prognosed NSV that is hospitalized versus receiving nothing

0.025 (0.01–0.05)

Assumed that prognosed NSV are unlikely to be incorrectly hospitalized

IMCI, integrated management of childhood illness; NSV, non-severe pneumonia; PO1, IMCI and pulse oximetry combination with a sensitivity of 70%; PO2, IMCI and pulse oximetry combination with a sensitivity of 85%; SV, severe pneumonia.