Table 3
Analysis of quantitative HIA studies: types of determinants, health outcomes and modelling, source of risk measures, and uncertainty
| Study | Proximal/distal determinants* | Type of health outcomes | Type of modelling in outcome assessment | Source of risk measures linking determinant to health | Time horizon | Uncertainty | |
|---|---|---|---|---|---|---|---|
| *Distal determinants affect health through intermediary factors. An example is income, which affects health via material circumstances, access to care, self esteem, etc. | |||||||
| 1 | Hallenbeck (1995): waste facility | Proximal | Life time cancer risks | Quantitative risk assessment (QRA) | Toxicological risk assessment based upon animal experiments | 70 years | Point estimates of maximum exposure only |
| 2 | Dunt (1995): highway | Proximal | Road traffic injuries | Traffic accident model | Local historical data on accidents by type of crossing | 10 years | Report includes sensitivity analysis |
| 3 | Zwart (1997): alcohol | Proximal | Number of excessive drinkers | Simple algorithm (Ledermann formula) | Cross country comparison | Unclear | Point estimate only |
| 4 | Mooy (1998): tobacco | Proximal | Disease-specific death rate, life years gained | Macro-simulation model with dynamic population (PREVENT) | Epidemiological evidence | 50 years | Point estimates only |
| 5 | Fehr (1999): waste facility | Proximal | Additional lifetime cancer risk, additional pop. cancer burden | Quantitative risk assessment (QRA) computer program | Toxicological risk assessment based upon animal experiments | 70 years | Point estimates only |
| 6 | Fehr (1999): highway | Proximal | Additional lifetime cancer risk, additional pop. cancer burden | Quantitative risk assessment (QRA) computer program | Toxicological risk assessment based upon animal experiments | 70 years | Point estimates only |
| 7 | Kemm (2000): home insulation | Proximal | Deaths per year | Innovative method | Ecological studies | Unclear | Point estimate dubbed “no more than a possible figure” |
| 8 | France (2000): freight distribution centre | Proximal | Mortality, morbidity, restricted activity days, hospital admissions. | Epidemiological risk assessment; three formulas for prediction of traffic accidents | Epidemiological studies; regional data. | Unclear | Air pollution: average and max. rates. Road traffic accidents: 3 point estimates. |
| 9 | Kemm (2000): woodprocessing plant | Proximal | Road traffic accident deaths and injury-only accidents | Analogy | National historical data on average number of accidents per distance travelled. | Unclear | Point estimate, uncertainty expressed by prefix “about”. |
| 10 | Kemm (2000): botanic garden | Proximal (accidents) and distal (income) | Road traffic accident injuries and deaths and %-age change in overall mortality (income) | Analogy (accidents); extrapolation from national data | National historical data on average number of accidents per distance travelled. | Unclear | Point estimates based upon “very uncertain assumptions”. |
| 11 | Taylor (2001): port extension | Proximal (air pollution, accidents) and distal (employment) | Respiratory deaths and hospital admissions; road traffic accident injuries; premature deaths (due to unemployment) | Epidemiological risk assessment (air pollution, employment); analogy (accidents). | Epidemiological studies (air pollution); local data (traffic accidents); single epidemiological (longitudinal) study (employment) | Unclear | Point estimates only (air pollution); terms like “3–4” and “<1” (traffic casualties); “just over three deaths” (employment). |
| 12 | Jobin (2003): Chad pipeline | Proximal | Deaths per year | Analogy (accidents, malaria), infectious disease model (HIV) | USA historical data (accidents); epidemiological study Nigeria (malaria), ? epidemiological data (HIV transmission rates) | Unclear | Point estimates of “likely” effects, intended to rank issues. |
| 13 | Fehr (2003): drinking water | Proximal | Additional lifetime cancer risk, additional cancer cases in pop. | Quantitative risk assessment (QRA) | Toxicological risk assessment based upon animal experiments (?) | 70 years | Point estimates and 5th and 95th centiles per increase of exposure (uncertainty in degree increase not quantified) |
| 14 | UK Dept. of Health (2001): foot and mouth disposal | Proximal | Additional variant Creutzfeldt-Jakob Disease infections as function of %-age of older cattle destructed. | Unclear | Unclear | Unclear | Point estimates with 95% confidence range. |
| 15 | Abrahams et al, (2002): foresight vehicle initiative | Proximal | Annual deaths and serious injuries (traffic accidents); first hospital admissions (air pollution) | Macro-simulation model with dynamic population (ARMADA)24 | Epidemiological evidence | 2000–2029 | Point estimates, uncertainty expressed by prefix “approximately” |
| 16 | Pitches and Kemm 2003): regional planning guidance transport chapter | Proximal | All-cause, cardiovascular and colon carcinoma deaths per year(s), acute myocardial infarction cases per year(s) (physical activity); slight, serious & fatal injuries (traffic accidents) | Epidemiological risk assessment (physical exercise); analogy (traffic accidents) | Epidemiological evidence (physical activity); national historical data on average number of serious personal accidents per distance travelled and mode of transport | Unclear | Mainly estimates in terms like “one death per 2 or 3 years”, and general word of caution: “estimates have very wide margin of error” |
