Table 1 Descriptive and methodological details of identified mask and respirator economic evaluations
First author, year, setting | Jones and Adida, 2013 [22], European contact rates | Chen and Liao, 2013 [21], Taiwan | Tracht et al, 2012 [11], USA | Dan et al, 2009 [12], Singapore | Cahill et al, 2008 [20], USA | Adal et al, 1994 [19], USA | Nettleman et al, 1994 [23], USA |
|---|---|---|---|---|---|---|---|
Infection(s) | Influenza epidemic | Seasonal influenza | Influenza A(H1N1)pdm09 | Influenza A(H1N1)pdm09, SARS, 1918 Spanish influenza | Influenza A(H1N1)pdm09 | TB | TB |
Mask(s) used | N95 | Surgical mask | N95 | N95 | N95, surgical mask | Isolation mask, respirators: DM, HEPA with/without disposable filter | Surgical cup mask, respirators: DM, DMF, HEPA |
Mask intervention | 1 mask/person/day for duration of epidemic (90Â days). Assumed respirator use begins when 0.05Â % population infected | Surgical mask use and natural ventilation | N95 respirator use by a varied % of the population for the duration of pandemic, starting when 0.001Â % symptomatic | Green 0: no intervention, Green 1: PPE for HCWs in contact with suspected cases, Yellow: full PPE for HCWs in high risk contact, Orange: PPE for HCWs in contact with medium risk | Monthly stockpiling and use for duration of pandemic | HCW program: respirators, fit testing and HCW medical evaluation | HCW program: 20 masks/8Â h shift for HCWs visiting patients in isolation |
Mask intervention effectiveness | Baseline effectiveness was 50Â %. Intervention estimated to reduce probability of infection to 30Â % or 70Â % of baseline, depending on person-to-person contact rates | Not explicitly reported | Intervention estimated to be 50Â % effective in decreasing susceptibility and 20Â % effective for reducing infectivity | Exposure reductions of 50, 80 and 90Â % with intervention. A 5Â % failure despite use of protective equipment and isolation measures | Probability of transmission in 5Â min encounter (varied for different % compliance for masks) | Not reported | Assumed respirator would prevent 25Â % of HCW exposure to TB |
Source of effectiveness data | Estimate derived from respirator assigned protective factor (APF = 10) [29]. This was adjusted for estimated lack of training, compliance and mask quality to give APF = 2 (i.e. mask 50 % effective) | Based on assumptions from previous study Chen et al. 2008 [33] where mask efficacies are assumed to be 60 %, 70 %, 80 %, or 95 % and are combined in the model with other control measures | Laboratory data, Lee et al. 2008 [52], and a randomised control trial by Aiello et al. 2010 [26] that found hand hygiene and facemask together were 35-51 % effective but not facemask use alone | No data cited for exposure reduction, these are assumptions. Failure rate estimate from a hospital simulation study Seet et al. 2009 [32] | Laboratory data from Balazy et al. 2006 [30] used to build particle transmission model | Reported none available | Reported none available |
Type of economic evaluation | Cost-effectiveness analysis | Cost-effectiveness analysis | Cost-effectiveness analysis | Cost-effectiveness analysis | Cost-effectiveness analysis | Cost-effectiveness analysis | Cost-effectiveness analysis |
Perspective | Policy developer view | Not stated | Not stated | Healthcare institution | Not stated | Not stated | Not stated |
Primary outcome measure | Total costs of intervention | Unit cost per person, per year | Net savings compared to no intervention | Incremental increase in cost per death averted | Productivity loss to economy from absenteeism | Cost of respirator use per case prevented and per life saved | Minimum estimates of cost per life saved and cost per death averted |
Intervention outcome measures | Cases | Cases | Cases, deaths, hospitalisations | Cases, deaths | Deaths, hospitalisation, outpatient visits, absenteeism | HCW PPD test conversion rates | Patients isolated for suspected TB, confirmed cases pulmonary TB in patients and active pulmonary TB in HCWs |