Women's injury in the home

Monash University Accident Research Centre - Report #158 - 1999

Authors: E. Cassell & J. Ozanne-Smith

Abstract

The major home injury problems have been identified for all levels of severity from an analysis of data from the Coroner’s Database (deaths), the Victorian Inpatient Minimum Database (hospital admissions) the Victorian Injury Surveillance System (hospital emergency department presentations) and the Extended Latrobe Valley Injury Surveillance project (a regional collection of data on General Practice presentations).

The higher-ranked causes of home injuries to adult women in Victoria are: falls (at all levels of severity); intentional-suicide and self harm (at all levels of severity except GP presentations); accidental poisoning (at the more serious levels of severity-deaths and hospital admission); and cutting/piercing injuries (at all levels of severity except fatalities). Middle-ranked causes at most levels of severity are: intentional-homicide and assaults; hit/struck/crushed; and fire/burns/scalds

The available data on these specific causes have been analysed and literature reviews conducted to inform recommendations for preventive action (strategies and countermeasures). For example, the recommended falls prevention measures include: promote fall-safe design to architects, draftsmen, builders and building suppliers; educate older women, health practitioners and carers about the importance of regular exercise, adequate nutrition (especially calcium intake), periodic medication review, treatment of osteoporosis and home fall hazard assessment and remediation; provide periodic falls risk assessment and tailored falls prevention programs including hip pads for high-risk older women; and develop and implement guidelines for slip resistive surfaces in bathrooms (including bath and shower bases) and kitchens and for outdoor pedestrian surfaces (including house entries and steps).

Executive Summary

Home injury is a neglected area of research. This may be because there is a general lack of appreciation of the size and gravity of the home injury problem. Also, the home is perceived to be a difficult setting to access for injury prevention purposes, particularly for adults, because it is part of the private domain. The Department of Human Services commissioned Monash University Accident Research Centre to investigate the size, nature and major causes of women’s injury in the home in Victoria and to make recommendations on strategies and countermeasures to reduce the burden of home injury.

Major findings

In Victoria, adult injuries are more likely to occur in the home (which includes the living space, garage and yard of a private dwelling, excluding the driveway) than in any other single location.
The home is the most frequently reported place of occurrence of injury to adult men and women (aged greater than or = 15 years) for hospital admissions, hospital emergency department presentations and general practitioner presentations.
According to death data used for this study (drawn from the Victorian Coroners’ Facilitation System - CFS) transport areas outrank the home for fatal injury but the CFS under-records elderly fall deaths due to fractured neck of femur. If Australian Bureau of Statistics (ABS) data on these deaths are included then the home outranks transport areas as the foremost location of injury fatalities.
At all levels of severity, including fatalities, approximately two-fifths (ranging from 37.8% for emergency department presentations to 45.1% for hospital admissions) of women’s injuries occur in the home. A similar proportion (39.5%) of men’s fatal injuries occur in the home but a lesser proportion, approximately one-quarter (ranging from 23.7% for GP presentations to 26.3% for hospital admissions) of men’s non-fatal injuries occur in the home.
By contrast, a greater proportion of males is injured in production and outdoor areas. These differences in location of injury between males and females are probably explained by differences in their exposure to particular locations.

Deaths

(Victorian Coroners’ Facilitation System – CFS)

According to the Victorian Coroners’ Facilitation System (CFS) data, the home is the second-highest ranked location of fatal injuries to adult women, behind transport locations. However, CFS data underestimates fall-related deaths. If the additional fall-related deaths in older women recorded by the Australian Bureau of Statistics (ABS) are included, then the home outranks transport as the location of most female injury deaths.
Approximately two-fifths (41.9%) of adult women’s injury deaths recorded on the CFS occur in the home. However, if ABS data on falls deaths are included, then home injury deaths account for over one-half (55.8%) of all injury deaths in adult females in Victoria.
On average, 271 women are fatally injured in the home each year (if ABS data on fall-related deaths are included).
According to CFS data, over one-half (55.6%) of reported home injury adult female deaths are intentional (suicide and homicide). Unintentional injury accounted for more than a quarter (28.7%) of deaths. However, if ABS home falls death data are included then unintentional injuries account for the majority (58.1%) of home injury deaths.
The all-cause home injury death rate for women aged 55 years and older was one-and-a-half times that for women aged 15-54 years. This differential would be much greater if ABS data on elderly fall deaths were included.

Hospital admissions

(Victorian Inpatient Minimum Database - VIMD)

The home is the highest-ranked location of injury requiring hospital admission.
Home injury accounts for 45.1% of adult female injury hospital admissions.
Women aged 60 years and older account for over two-thirds (67.9%) of hospital admissions for home injury.
Falls cause over two-thirds (68.7%) of home injury hospital admissions. The other major cause of admissions is intentional self-inflicted injury.
Fractures are the most frequently reported injury accounting for nearly one-half (47.0%) of the home injury admissions.
Older women are more at risk of admission for fractures, open wounds, bruises and burns, younger women are more at risk of hospitalisation for self-poisoning (intentional and accidental).

Hospital emergency department presentations

(Victorian Injury Surveillance System - VISS)

Over one-third (37.8%) of women presenting with injuries to the emergency departments of VISS hospitals are injured in the home.
Unintentional injury accounts for over four-fifths (84.9%) of VISS home injury presentations.
The major causes of home injury are falls (37.5%) and cutting and piercing (16.4%).
Most home injuries occur in the living and sleeping areas (51.2%) and in the garden, garage and yard (27.0%).
The five most common non-systemic injuries are: cuts and lacerations (21.0%), fractures (18.4%), bruising (10.3%), inflammation/swelling/pain (10%) and sprain or strain (9.9%).
The five activity groupings most associated with VISS presentations for home injury is: leisure and recreation (excluding sports), miscellaneous household activities, "other" catastrophes, home maintenance and personal activities.
The five most hazardous individual activities are: "other" leisure/recreation activities, intended self-harm, fight, riot or quarrel, playing (general activity) and cleaning.
The five environmental factor groupings that are most frequently reported as precipitating injuries to women in the home are (in rank order):

their own or another person's actions;

structures (particularly stairs and steps and floors and flooring materials, mostly in relation to fall injuries);

drugs and medications (particularly the sedative/tranquillisers/psychotropic group) and alcohol (related to deliberate self-harm and accidental overdoses);

furniture (particularly chairs, beds, bathtubs and showers, mostly related to fall injuries); and

Kitchenware (particularly knives and drinking glasses, mostly related to cuts).

General Practitioner (GP) presentations

(Extended La Trobe Valley Injury Surveillance – ELVIS)

Over two-fifths (42.8%) of adult women presenting with injuries to ELVIS GPs are injured in the home.
Almost all (97.7%) of the home injury presentations to GPs are for unintentional injuries (although intentional injuries may be under-reported and undetected).
The most frequently occurring injuries are: lacerations (25.0%), sprains and strains (19.5%), bruising (14.9%), bites (10.0%) and burns (6.3%).
Falls and cutting/piercing are the highest-ranked causes of GP presentations each accounting for approximately one-quarter of the home injury presentations.
The peak times for home injuries presenting to GPs are from Saturday to Wednesday (inclusive) and in the summer months, January to March.
Women aged 60 years and older appear more at risk of injury. They comprised 19.5% of the Latrobe Valley Region population in 1994-95 yet accounted for 39.4% of all adult female injury presentations to GPs over the same period.

Major causes of injury across all levels of severity

The higher-ranked causes of home injury to adult women are:

falls (at all levels of severity if the underestimation on the CFS of fall deaths among elderly women due to fractured neck of femur is taken into account)
suicide and self inflicted injury (at all levels of severity except General Practitioner (GP) presentations)
accidental poisoning (at the more serious levels of severity - deaths and hospital admissions)
cutting and piercing injury (at all levels of severity except deaths)

Mid-ranked causes of home injury across most levels of severity in Victoria are:

hit, struck and crushing injuries
fire, burns and scalds
intentional injury - homicide and assaults

The lower-ranked causes are natural and environmental injury (mostly animal-related injury), choking, suffocation and foreign body in orifice and over-exertion injury. The findings and recommendations on these causes included in the report proper are not covered in this summary.

Major causes in detail

1. Falls

Falls are the leading cause of injury to adult women in the home at all levels of severity except fatalities. However, fall-related deaths are underestimated on the Victorian Coroners’ Facilitation System because fall-related deaths of older women due to fractured neck of femur are not routinely recorded.

Falls accounted for:

five per cent of home injury deaths recorded on the CFS (this is a gross underestimation because elderly fractured neck of femur (hip) fall deaths is not included on the CFS)
over two-thirds (68.7%) of VIMD home injury hospital admissions;
over one-third (37.5%) of home injury VISS hospital emergency department presentations; and
approximately one-quarter (26. 0%) of ELVIS General Practitioner presentations for home injury.

Overall, fall injuries are mostly caused by slips, trips and stumbles on the same level; same-level falls caused between approximately one-half and three-quarters of fall injuries at every level of severity.

The falls fatalities recorded on the CFS are more likely to be caused by falling from one level to another than from falling on the same level. The data are biased because elderly fractured neck of femur fall deaths are not recorded on the CFS and are generally reported to be same-level falls.

Women aged 60 years and older appeared to be at higher risk of fall injuries at all levels of severity, including fatalities. Risk also appears to increase with age, from age 60 years onwards.

Type of injury

Head injuries are the most frequently reported cause of fall deaths on the CFS database (ABS data shows fractured neck of femur to be the most common cause of fall-related deaths).
Fractures account for over two-thirds (68.3%) of VIMD hospital admissions, nearly one-half (48.3%) of VISS emergency department attendances and 16.2% of ELVIS GP presentations.
Other frequently reported injury groupings are sprains and strains (for emergency department and GP presentations), cuts and lacerations and bruising.

Falls on the same level

The analysis of a sample of VISS one-line case narratives on hospital emergency department presentations (admissions and non-admissions) and all ELVIS GP case narratives reveals a common pattern of causes.

The prominent causes of same-level falls are: loss of balance, dizziness or fits, body joint ‘giving way’ (hip, knee or ankle); slips on wet or icy surfaces (indoor and outdoor); and trips involving a broad range of items.
The slipping surface is described as wet or icy in over one-third of the VISS and ELVIS case narratives but this is probably an underestimation as the quality of the narratives varied. In fact, it is generally accepted that slips typically occur on wet surfaces or a surfaces contaminated with oil or other spills.
The most frequently reported tripping hazards are: furniture (chairs and beds), steps and stairs, mats, uneven concrete paths, garden surrounds, hoses, cords and animals. These are mostly portable/moveable items which highlights the importance of educating householders to keep indoor and outdoor pathways (especially to the clothesline and car) free of clutter and protruding furniture.

Falls from one level to another

Falls from one level to another account for just over one-half of CFS fatalities, one-third of VISS hospital emergency department and ELVIS GP presentations and a significant proportion of hospital admissions (15.8%).

At all levels of severity, except hospital admissions, the majority of falls are on indoor and outdoor stairs and steps and from chairs or beds.
Steps and stairs are involved in one-third of hospital admissions for different level falls. Research shows that high-risk stairs include those with fewer steps, low risers, treads less than twelve inches deep, abrupt changes of conditions and distracting surrounding vistas. Poor lighting and distracting patterned carpet are particular problems for the elderly.
Falls from ladders are also a prominent cause of death and injury. The analysis of VISS narratives revealed that using stools and chairs as climbing apparatus when doing household chores was hazardous and that this type of fall was over-represented in VISS hospital admissions in proportion to emergency department presentations.

Recommendations

Countermeasures and strategies

Continue and expand the education of architects, draftsmen, designers, builders and manufacturers and suppliers of building materials and home appliances and equipment on the safe design of homes using A/NZ 4226-1994 Guidelines for safe housing design and other relevant standards as a basis for this education.
(In partnership with general practitioners, other health professionals and aged care agencies) systematically expand primary and secondary prevention strategies and countermeasures to reduce falls among older women by providing:

education (promotion) and counselling about the importance of regular exercise, adequate nutrition and the identification and remediation of home fall hazards;

periodic review of medications that increase fall risk;

education about the potential benefits and risks of HRT and other drugs used to prevent and treat osteoporosis;

periodic assessment of at-risk older women, and the institution of tailored remedial action on identified predisposing and behavioural risk factors for falls and fall injuries; and

supportive environments to enable strategies and actions, which promote falls prevention and health maintenance to occur and be maintained.

Promote the wearing of protective hip pads to women at high risk of fall injury in both the community and aged care settings.
Promote the use of proven slip resistive surfaces in bathrooms (including bath and shower bases), kitchens and all outdoor pedestrian areas (including entries and steps).
Widely promote a set of simple guidelines to prevent home fall injuries.

Surveillance, research and investigations

Improve the reporting of place of occurrence (location) of injury in surveillance systems (particularly VIMD).
Develop simple guidelines to standardise information on the circumstances and contributory factors to falls in the one-line case narratives in surveillance systems to increase their usefulness.
Undertake an annual reconciliation of the Victorian home fall deaths data recorded on the Coroners’ Facilitation System (CFS) and Australian Bureau of Statistics database.
Conduct follow-up telephone surveys on injury cases involving slips, trips and stairway/step falls reporting to hospital emergency departments (through VEMD) to provide more information on exposure and the intrinsic, behavioural and environmental factors contributing to injurious falls.
Advocate for and support the independent testing (under wet conditions) of household pedestrian surface materials advertised as slip-resistant or products advertised as conferring slip resistance to existing indoor and outdoor surfaces (in partnership with CSIRO and the Australian Consumers’ Association).
Support research to underpin the development of guidelines (and ultimately an Australian Standard) on the use of slip-resistive pedestrian surfaces for private dwellings
Conduct a controlled study in institutional settings for older people with good falls recording systems to evaluate the effectiveness of slip-resistant flooring and floor treatments as falls prevention measures.
Conduct research to more precisely describe the role and contribution of environmental factors to falls among older women and to better describe the mechanisms by which environmental factors contribute to falls.

2. Suicide and self-inflicted injury

Intentional self-inflicted injury is a high-ranked cause of home injury among adult women at all levels of severity, except for GP presentations.

Intentional self-inflicted injury accounted for:

approximately one-half (47.0%) of home injury deaths reported on the CFS; and
a significant proportion of VIMD hospital admissions (7.1%) and VISS emergency department presentations (9.1%).

Women aged 45 years and older are more at risk of suicide. Women aged under 40 years appear to be the highest risk group for both VIMD hospital admissions and VISS emergency department presentations for self inflicted injury, accounting for approximately three-quarters of cases on both these databases.

Causes (means) of suicide

poisoning by solid and liquid substances (45.2% of suicide cases) mostly by "other specified drugs and medicinal substances";
hanging, strangulation and suffocation (29.9%) mostly by hanging or suffocation by plastic bags; and
other gases and vapours (12.6%) almost all by car exhaust (CO) gas.

Causes (means) of self inflicted injury

Poisoning by solid and liquid substances: 92.4% (VIMD hospital admissions) and 84.9% (VISS hospital ED presentations), mostly by pharmaceuticals; and
cutting/piercing: 5.5% (VIMD hospital admissions) and 11.1% (VISS hospital ED presentations).

Self-poisoning by pharmaceutical drug overdose is the foremost cause of suicides, hospital admissions and VISS emergency department presentations for self inflicted home injury among women. Prescription drugs predominate. This pattern is also evident in accidental poisoning. The categorisation of pharmaceutical poisoning injury cases as intentional or accidental involves some subjectivity where evidence of intent is equivocal or unavailable. Therefore, intentional and unintentional self poisoning by pharmaceutical drugs are discussed together in chapter 9. Recommendations for prevention focus on measures that have good potential to reduce community access to those drugs that are over-represented in intentional and accidental self-harm cases with some consideration of data on prescriptions.

Drugs implicated in self poisoning (suicide and accidental poisoning) deaths

Information on the drug agents implicated in deaths was only available for 17.0% of the drug-related suicides and 31.6% of drug-related accidental poisoning on the CFS database. Therefore, the findings reported here are tentative because the sample may not be representative and the number of specific drugs in each grouping was small.

When considered together, the pharmaceutical drugs most implicated in suicide and accidental poisoning deaths are:

tricyclic antidepressants, particularly amitriptyline, doxepin (for suicides) and dotheipin (more prominently in accidental poisoning);
opioid analgesics, particularly morphine/heroin (for suicides) and methadone (for accidental poisoning);
benzodiazepine-based hypnotics and sedatives particularly flunitrazepam and temazepam; and
anxiolytics (particularly diazepam) for accidental poisoning deaths.

Drug involved in hospitalisations (intentional and accidental)

Intentional poisoning (suicide attempts)

The drug groups most frequently involved in VIMD hospital admissions for intentional self inflicted poisoning are:

tranquillisers and other psychotropics, including antidepressants (56.6%); and
analgesics, including heroin and other opiates (18.1%).

The number of hospital admissions with heroin involvement could not be disaggregated from other analgesics for self inflicted injury cases on the VIMD. VISS hospital admissions data suggest that heroin-related admissions for self-inflicted injury are few and that the specific drug most involved in analgesic overdoses is paracetamol.

The specific prescription drugs most implicated in self-inflicted poisoning in VISS hospital admissions were diazepam, temazepam and dothiepin.

Accidental poisoning

Ninety per cent of all accidental self-poisoning VIMD hospital admissions involved prescription drugs, mostly predominantly benzodiazepine-based tranquillisers (21.0%), anti-depressants (19.3%) and paracetamol-based analgesics (16.2%). Heroin, methadone and other opiates were reportedly involved in only 3.0% of the accidental poisoning hospitalisations.

Prevention and control of self poisoning

Efforts to reduce prescription drug-related deaths and hospital admissions should concentrate on primary and secondary prevention in the community including measures to:

reduce community access to the drugs that are most involved in self-poisoning;
ban drugs that have little therapeutic value and are implicated in self poisoning deaths (such as chloral hydrate and colchicine); and
restrict the availability of drugs that are over-represented in deaths and admissions in proportion to prescription frequency

Recommendations

Establish a national poisoning prevention and control advisory committee comprised of clinicians researchers, industry representatives and other interested parties to:

monitor trends in mortality and morbidity related to toxic substances (both prescription and non-prescription), along with changing prescribing habits; and

make recommendations on reducing the availability and scheduling of identified hazardous substances (Coleridge et al 1992, Buckley et al 1995b, Cantor et al 1996).

Support regulations to further restrict the availability of barbiturates (particularly pentobarbitone) and dextropopoxyphene or have them withdrawn from the market (Myers et al 1981, Cantor et al 1989, Buckley et al 1995b, Cantor et al 1996).
Limit the prescription size of anticonvulsants available on the PBS for epileptics with co-existent psychiatric problems (Buckley et al 1995b).
Monitor the involvement in self poisoning deaths of the ‘newer’ and reportedly less toxic antidepressants such as fluoxetine and mianserin and, if they prove less toxic than ‘older tricyclics’ (and as efficacious), promote their preferred use in the treatment of depression (Rettersol 1993; Gunnell and Frankel 1994; Malmvick et al 1994; Henry et al 1995, Buckley et al, 1995a, Cantor et al 1996).
Negotiate guidelines with the media to get them to avoid drawing attention to specific methods when they report suicides to prevent imitative suicides among young people (Cantor et al 1996).
Support current negotiations being conducted by a coalition of key stakeholders and convened by the Federal Office of Road Safety and the Australian Medical Association to ensure the implementation of measures to reduce motor vehicle exhaust gassing suicides including engineering solutions or performance requirements which limit carbon monoxide (CO) emissions in car exhaust systems or prevent access to lethal doses of CO (Gunnell & Frankel 1994, Cantor et al 1996, Routley 1997; Routley 1998).
Initiate or promote programs which enable the community to safely and conveniently dispose of unwanted or expired medication based on the successful program in the Hunter Valley region of NSW (Cantor et al 1996).
Educate prescribers to consider the toxicity of drugs in overdose when making decisions on treatment and to more actively monitor compliance with treatment especially for patients with depression or experiencing traumatic life events, for example, the death of a family member or marital separation.
Educate treatment agencies and general practitioners about the need for caution and careful monitoring when prescribing methadone and benzodiazepines to drug addicts (Coleridge et al 1992).
Consider the findings and recommendations from research into factors contributing to the rise in suicides by hanging in Australia, currently being undertaken by the Australian Institute for Suicide Research and Prevention

Surveillance, research and investigations

Support the establishment of a national coronial database to better identify trends in suicide, the methods used and contributory factors, including specialised modules that focus on suicide and limit illicit drugs.
Routinely link toxicological data to the National Coronial Information System.
Standardise definitions to guide the classification of self-poisoning cases by intent, ie., self inflicted, accidental and undetermined, in all databases.
Standardise methods of determining cause of death from poisoning and methods for attribution of aetiological fractions (Buckley et al 1995b).
Improve the recording of place of occurrence (location) of self-inflicted injury on surveillance systems, particularly VIMD hospitalisations.
Develop simple guidelines to standardise information in one-line case narratives in surveillance systems to increase their usefulness.
Standardise ICD classification for heroin-related deaths and hospitalisations (self-inflicted and accidental) and re-classify heroin-overdose deaths that are apparently wrongly classified under ‘adverse effects of drugs in therapeutic use’ in the CFS.
Support research to determine the mortality and morbidity caused by paracetamol overdose in Victoria including a closer examination of the trend of its use in self poisoning over time, and on the cost-benefit of potential countermeasures, for example, limiting availability and inclusion of an antidote (Hawton et al 1996; Gunnell & Frankel 1994; Gazzard 1976, Cantor et al 1996).

3. Poisoning (accidental)

Non-intentional (accidental) poisoning is a major cause of more serious home injuries—fatalities and hospital admissions.

Pharmaceutical drug overdoses

Pharmaceutical drug overdoses are largely responsible for the fatal and serious non-fatal accidental poisoning. Although data are limited, the most frequently used classes of drugs involved in accidental poisoning appear to be: psycholeptic benzodiazepine-derived anxiolytics, sedatives and hypnotics; antidepressants; and analgesics.

Alcohol

Alcohol (taken by itself) is the most prominent of the other accidental poisoning agents. It was reported as the primary cause of 16.7% of accidental poisoning fatalities and 10.1% of accidental poisoning VISS emergency department presentations.
Additional information was only given in one-third of the case narratives on fatalities and in all of these cases the victim was described as an alcoholic or a person with a history of alcohol abuse. Information in the VISS narratives is also limited. The circumstances reported in one-half of the cases involving alcohol are: binge drinking at parties; drinking in response to domestic problems and feelings of depression; and a history of alcohol abuse.

Foodstuffs and plants

A range of foodstuffs and poisonous plants (most frequently mushrooms and other fungi) are implicated in a small proportion of VIMD hospital admissions (3.0%) and VISS emergency department presentations (5%) for accidental poisoning.

Other agents

Cleansing and polishing agents contribute to a significant proportion of less serious injuries (8.1% of VISS emergency department presentations). In a small number of VISS cases the victims mistook chemical cleaning substances for beverages (Kemdex solution for milk, lemon scented bleach for lemon drink, carpet cleaner for lime soda, industrial window cleaner stored in a lemonade bottle for lemonade, oleander fluid for wine).

Age factors

Women aged 40 years or older appear at higher risk of accidental poisoning death whereas younger women appear to be at higher risk of hospital admission and emergency department presentation for accidental poisoning. The same pattern is evident for suicide and self inflicted injury and requires further investigation.

Other contributory factors

Contributory factors to accidental poisoning are not well or consistently reported in the one-line narratives in the Coroners’ Facilitation System but a history of drug abuse or chronic mental or physical illnesses appeared to be predisposing factors. Just over one-half of the accidental poisoning cases on the VISS databases appeared to be deliberate self poisoning with drugs and alcohol and highlight the difficulty of assigning self poisoning cases to intentional or accidental injury classifications.

Recommendations

The recommendations relating to accidental poisoning by pharmaceutical drug overdose are in the suicide and self-inflicted injury summary above.

Strategies and countermeasures

Discourage manufacturers of cleaning products from adding perfumes and colours to cleaning agents that increase the risk of users (especially elderly people) confusing cleaning products with beverages.
Educate householders to avoid decanting cleaning agents into drink bottles and to store all cleaning agents securely, away from sinks and separate from areas used to store bottles of drink.

Surveillance, research and investigations

Develop definitions (and guidelines for their application) for accidental and self-inflicted poisoning to assist the consistent classification of cases in surveillance databases.
Improve the content of one-line narratives in injury surveillance systems so that the specific poisoning agents (including BAC reading for alcohol-related deaths), circumstances, predisposing and contributory factors are consistently reported.
Investigate the different age-related pattern for self-poisoning deaths (intentional and accidental) compared to hospitalisations and emergency department presentations for self poisoning.

4. Cutting and piercing injury

Cutting and piercing injuries caused only two deaths in the five years covered by the CFS database. However, cutting and piercing injuries are the third-highest ranked cause of hospital admissions, and the second-highest ranked cause of both hospital emergency department presentations and GP presentations.

The specific products most involved in cutting and piercing injuries at all levels of severity, except deaths, are: knives, ‘other (non-powered) hand tools and implements’ and glass.
Lawn mowers cause a significant proportion of the cutting and piercing injuries that required hospital admission.

Knife cuts

Knives are involved in approximately one-quarter of VIMD hospital admissions for cutting and piercing injury and ELVIS GP presentations (25.4% and 27.3% respectively) and three-tenths (30.8%) of VISS emergency department presentations.
The analysis of VISS emergency department and ELVIS GP presentations data revealed that 70%-80% of knife cuts occur during food preparation and a comparatively smaller proportion (6%-11%) occur when knives were being washed or dried.
The most common classes of foods being cut at the time of injury are fruit and vegetables and meat. The specific foods most reported as involved in knife cutting incidents are: meat (not specified), vegetables (not specified), cake, pumpkin, potatoes, carrots, leg of lamb and opening oysters.
Details of the circumstances of the injury and food being cut at the time of injury are inconsistently reported in VISS emergency department and ELVIS GP case narratives.

Recommendations

Strategies/countermeasures

Upgrade education about knife use and safety in post-primary Home Economics curriculum especially at the junior secondary school level.
Develop and conduct an injury prevention campaign consisting of radio messages about the selection, safe use and storage of kitchen knives (linked by a 008 telephone number to a mailout of a brochure on knife use and safety).

Guidelines on the selection, safe use and storage of knives are included in the report proper.

Research/investigations

Conduct a follow-up research study of knife cut cases presenting to hospital emergency departments to more precisely determine:

the circumstances of knife cuts; the relative contribution of causal factors (for example, design of knife, sharpness of blade, incorrect use, poor cutting technique, sight and hand/finger disabilities);

the nature of the injury (including which hand was injured);

the food or item being cut at the time of injury; and

consumer acceptance of potential countermeasures (including protective gloves).

Investigate the suitability (including hygiene issues) and acceptability to consumers of wearing a protective gloves (or one glove on the non-dominant hand) when cutting food.
Investigate the feasibility and practicability of introducing an Australian and New Zealand Standard for Hand-held knives for use in home food preparation based on Standard AS 2336-1992 Meat industry-Hand-held knives.

Sewing needles and pins

‘Other hand tools and implements’ account for 9.9% of hospital admissions for cutting and piercing injury in the VIMD database and a similar proportion of VISS emergency department and ELVIS GP presentations (8.7% and 11.7% respectively).

No finer breakdown codes are available to identify the specific hand tools and implements involved in hospital admissions. However, the analysis of VISS emergency department data (for both admitted and non-admitted cases) revealed that needles and pins were the most frequent cause of injuries in this classification and the injuries mostly occurred when sewing needles and pins, predominantly the former, were trodden on. These are serious injuries; 90% of the ‘trodden on’ piercing cases required hospital admission, presumably to remove the needle or pin. Needle and pin pierces were a less prominent cause of GP presentations, ranking behind scissor cuts.

Recommendations

Strategies and countermeasures

Train students learning garment construction, tailoring and hobby crafts and people practising these crafts to routinely ‘sweep’ the activity area with a magnet after each session involving pins and needles.
Encourage manufacturers to include magnets in hand sewing and sewing machine accessory kits.
Advocate for the inclusion of safety tips/hazard warnings about needle and pin injuries in sewing machine instruction booklets and on needle and pin packets.

Broken glass

Information on the frequency of broken glass-related VIMD hospital admissions are not available because these injuries are classified with other causes in the ICD9 system under ‘other specified cutting and piercing instruments’. However, VISS hospital admissions data give some indication of their frequency.

Broken glass injuries accounted for two deaths, just over one-quarter (26.1%) of all VISS hospital admissions for cutting and piercing injuries, 21.1% of all VISS cutting and piercing non-admissions and approximately one-tenth (9.8%) of cutting and piercing injuries recorded on the ELVIS GP database.

One of the two deaths and between one-quarter and one-third of the glass-related cutting and piercing injuries resulting in VISS emergency department and ELVIS GP presentation are caused by broken glass from windows and doors. The most common scenarios reported in VISS and ELVIS case narratives were ‘tripped fell through glass door/window’, ‘put hand through door/window’, ‘knocking on door/window, glass broke’.
Other prominent causes of glass cuts are: ‘broke glass when washing dishes’, ‘broke glass during cleaning tasks/handling’, ‘stepped on broken glass’.

In September 1991 Victoria incorporated the 1989 revision of the Australian Standard (AS1288-1989 Glass in Buildings - Selection and Installation) into Victoria's Building Code. This mandated the use of safety glazing materials (toughened, laminated or organic) in some residential ‘high risk’ situations in new housing or houses undergoing renovations that require a permit. Hazardous glass in existing housing stock is still a problem. There is no onus on householders to replace broken annealed glass with safety glass.

Recommendations

Strategies and countermeasures

Educate suppliers, glaziers and insurance companies to recommend the replacement of annealed glass with safety glass whenever broken glass is replaced in the doors and windows of homes.
Promote the advantages of safety glass to consumers.
Reduce the price difference between safety and annealed glass.
Educate householders to:

apply special plastic film, bars, rails, warning stickers or decals on hazardous glass in the home; and

select safety glass when broken glass is being replaced in windows and doors.

Surveillance, investigations and research

Investigate whether the incorporation of the 1989 revision of the Australian Standard (AS1288-1989 Glass in Buildings - Selection and Installation) into Victoria's Building Code regulations has been effective in reducing architectural glass injuries in the home and other high risk settings.

Lawn mower cuts

Lawn mower injuries accounted for 8.0% of hospital admissions for cutting and piercing home injuries on the VIMD database and a lesser proportion (3.7%) of VISS hospital emergency department presentations (but were over-represented in VISS admissions compared to other causes of cutting and piercing injuries).
Lawn mower injury accounted for only a small proportion (1.9%) of the cutting and piercing injuries on the ELVIS General Practitioner database.
The analysis of the case narratives of a sample of VISS emergency department presentations revealed that most injuries occurred when the user’s hand or foot came into contact with the blade of the mower.
The hand injuries were caused when the user put her hand too close to the blades when taking the grass catcher off, adjusting the blades (with motor running) and removing blocked grass. These injuries usually required hospitalisation.
The foot injuries occurred when the user’s foot slipped under the mower while it was operating.

Note that injuries to the eye caused by objects thrown up by mowers are classified under a different ICD9 E-code: ‘Accidents caused by submersion, suffocation and foreign bodies’.

Recommendations

Strategies and countermeasures

Review the design of powered hand-held motor mowers to provide a discharge chute that deflects struck objects in a downward direction.
Revise the voluntary standard AS/NZS 2657-1985 Powered Rotary Lawnmowers to require:

a prominent label (visible from operating position) that warns of the danger of hand and feet injuries (including time of blade run-on); and

an operator presence control (OPC) design feature that shuts off the blades if the operator leaves the normal operating position.

If research findings support regulatory action mandate that all powered mowers sold in Australia meet the appropriate (revised) AS/NZS standard for lawnmowers (AS/NZS 3792-1992; AS/NZS 2657-1985).
Raise consumer awareness of the hazards of lawn mower use and the need to:

use the safety features provided consistently and wear protective clothing (eye protection, boots, gloves and ear muffs);

keep hands clear of blades whenever the engine is running and/or the blades are moving;

shut down mower engine and wait for blades to stop circling before adjusting the mower or its parts in any way, unclogging the blades or blade area or carrying it;

remove stones and other debris from the area before commencing mowing and keep bystanders clear from possible flying objects;

refuel mower out-of-doors and well away from open fires and cigarettes; and

use a portable safety switch when using an electric mower and to take special care never to use an electric mower near water or when it is raining and never to pull it backwards.

Surveillance, research and investigations

Conduct a telephone follow-up survey of cases presenting to emergency departments of VEMD hospitals with hand, foot, eye and other injuries associated with lawn mowers to more precisely determine the causes of these injuries, in particular the relative contribution of design factors and unsafe user behaviour. Use the findings to revise the Australian and New Zealand Standards for lawnmowers with a view to mandating the Standard if the survey shows this step to be warranted.

5. Hit, struck and crushed injury

Hit/struck/crushed injuries are rarely fatal, only two deaths are recorded on the CFS over the five-year covered by the database. Although these injuries are a middle-ranked cause of VIMD hospital admissions and VISS emergency department presentations, they account for only a small proportion of home injuries at each of these levels (2.4% and 5.5% respectively). They are a more prominent cause of minor injuries, accounting for 16.2% of ELVIS GP presentations.

At all levels of severity, except fatalities, the most frequent cause of hit/struck/crushed injuries is victims striking against or being struck by objects or persons.
Objects falling on the victim caused the two fatalities. Falling objects were also prominently involved in VISS emergency department presentations, causing approximately one-third of injuries at that level of severity.

It is estimated from available data that 190 women are admitted to hospital each year in Victoria for hit/struck/crushed injuries that occurred in the home. The annual frequency of emergency department and GP presentations cannot be estimated from the available data, as they are not complete. Older women (women aged 60 years and older) appear to be more at risk of hit/struck/crush injuries at all levels of severity

Hit by or struck against injuries

Overall, most hit/struck/crushed injuries occurred when the victim is hit by or struck against an object when doing chores around the house and in the yard and garden.
Detailed information on injury events is not available for VIMD hospitalisations. The analyses of the CFS, VISS and ELVIS case narratives revealed that a large number of objects are implicated. For example, there were 26 different objects involved in the 56 ‘striking against/struck by’ injury cases on the VISS emergency department database (excluding cases involving persons and animals).
Doors (hitting victims), motor mowers or whipper snippers (kicking up debris onto users and bystanders) and hammers (mainly hitting the user’s finger or hand) were the objects most frequently involved in VISS injury cases (although the number of reports in each grouping was small). Items of furniture (stools, chairs, coffee tables, beds, couches, cupboards and benches) and doors were the objects most frequently involved in ‘striking against/struck by’ injury cases presenting to GPs.

Struck by falling objects

The two deaths recorded on the CFS were each caused by falling objects: a tree limb and a wardrobe.

Household furniture items (beds, drawers, tables, wardrobes and cupboards and chairs and stools) are the most prominent of the ‘struck by falling object’ cases that presented to VISS hospital emergency departments and ELVIS General Practitioners. The heavier furniture items caused injury when they were being moved.

Crushing injuries

Door jam injuries (involving both swing and sliding doors) are the most frequent ‘caught between or in’ injuries. The analysis of VISS hospital emergency department narratives showed that a variety of doors were involved e.g. room, oven, cupboard, shower and car. In ELVIS GP narratives room and entry doors were most commonly implicated.
The closing, not the hinge, side of household doors appears to be the problem for adults (in contrast to children’s door jam injuries).

Type of injuries

Open wounds and fractures are the most frequent causes of hospital admissions. Bruising and cuts and lacerations (to the foot, toe, finger, hand and face) predominate in VISS emergency department and ELVIS General Practitioner presentations.

Recommendations

Strategies and countermeasures

Educate householders to implement the following safety measures:

Install door stops and door closures on room and entry doors to prevent door-jam injuries.
Fix heavy wardrobes, bookshelves and wall units to walls.
Only move heavy items (wardrobes, wall units and shelving systems) if trained, and use special moving equipment as appropriate
To prevent injuries from debris thrown up by lawnmowers:

clear area of all debris before commencing mowing

allow no bystanders or helpers (including people gardening) around the area during mowing

wear safety equipment (goggles, ear muffs, gloves, heavy clothing and boots).

(see also recommendations to prevent lawnmower-related cutting and piercing injuries)

To prevent hammer-related injuries:

wear goggles and gloves

make sure hammer head is not chipped or burred

strike a hammer blow squarely, with the striking face of the hammer parallel to the surface being struck

use only the striking face of the hammer, never the side or flat of the hammer.

6. Fire, burns and scalds

Fire, burns and scald injury is a middle-ranked cause of adult female home injury at all levels of severity. On average, between 9 and 10 adult women die each year in Victoria from fire, burns and scalds that occur in the home (representing 6.0% of injury deaths recorded on the CFS). Projected from available data, an estimated 150 women are admitted to hospital with fire, burns and scald injuries each year (2.0% of VIMD hospital admissions for injury). Burns and scalds account for 4.3% of VISS emergency department presentations and 5.9% of GP presentations. The annual incidence of hospital emergency department and GP presentations cannot be estimated from the available data because they are not complete.

A different pattern of causes was evident for fatalities when compared with fire, burns and scald injuries at the other levels of severity.

Over four-fifths of the deaths are caused by fire and flames.
Scald injuries from hot liquids and vapours, including steam, predominate at the other levels of severity (hospitalisations, emergency department presentations and GP presentations), where they contribute approximately one-half of the fire, burns and scalds cases. However, fire and flame injuries are still a significant cause of hospitalisations, contributing one-third of VIMD hospital admissions.
Burns from hot objects are less frequent overall, but contribute a significant proportion of hospitalisations and less severe injuries (14% and 29% respectively).

These major causes will be discussed separately because, in general, they require different preventive interventions.

Burns and smoke inhalation associated with fire and flames

Deaths

House and room fires and ignition of clothing and bedclothes cause most deaths due to burns and smoke inhalation.
The predominant sources of ignition for fatal fires are smoking products (unextinguished cigarettes or matches) and faulty electricals either house wiring or household items, particularly electric blankets.
The ignition sources of fatal fires are different for the older and younger age groups. Smoking-related products are the predominant ignition source for the fires involving the younger age group (women aged under 65 years), followed by faulty electrical products and wiring. By contrast, only one of the eleven fires started by smoking-related products occurred in the older age group (women aged over 65 years).
The main sources of ignition for the fatal fires involving older women (who were the highest risk group for fire-related deaths) are:

heating sources (radiators, pot belly stove, open fire);

cooking stoves (either fires from food left unattended, gas left on or clothing catching fire while - the victim was cooking); and

faulty electric products or wiring.

Electric blankets are the electrical product most implicated in fire-related deaths for both the younger and older age groups, although a possible alternative ignition source was reported for three of the five fatal fires attributed to electric blankets.
Alcohol is mentioned as a factor in three fire-related deaths, but alcohol involvement is probably inconsistently reported.

Hospital admissions

The main causes of VIMD hospital admissions for fire and flame burns are: house and outbuilding conflagrations; ignition of clothing; ignition of highly inflammable material (fuels, fat) with ignition of clothing; and burning by controlled fires (sources of heating).
The VIMD does not give any more information on sources of ignition and other contributory factors, and the VISS one-line narratives were not useful in establishing causal patterns as there was only a small number of fire and flame hospitalisations on the VISS database. Data from fire authorities reveal that the type of material most frequently ignited in house fires (defined as fires in one- and two-family dwellings) is food fat, grease or oil.
Women aged 60 years and older are most at risk of hospital admission for burns.

Scalds by hot liquids and vapours

Six of the seven scald injury deaths recorded on CFS were caused when the predominantly elderly victims were bathing or showering (the other was caused by a spilt pot of hot soup). Epilepsy or ‘fitting’ was reported as a factor in three of the six scald deaths that were associated with bathing.
Over one-half of VIMD hospital admissions are for scalds from hot liquids and vapours but the E-coding does not differentiate between types of liquids and steam. VISS emergency department data for admissions suggest that hot water (predominantly contacted when bathing) and hot oil splashes and spills are the main causes of the more severe burns and scalds, although the number of cases in the VISS hospital admissions subset was small (n=31).
Hot water scalds are the primary cause of VISS emergency department non-admissions. These injuries are predominantly caused by boiling or hot water spills that occur when victims are making hot drinks or cooking (less than 1.0% occurred during bathing). This pattern was also evident in GP presentations. Oil and fat splashes and spills during cooking, hot drink spills and steam burns are the other major causes of less severe burn and scald injuries (VISS hospital emergency department non-admissions and G.P presentations).

Burns from hot objects

There were no deaths attributed to burning by hot objects on the CFS, although there was one case where a woman collapsed against a heater, which set her clothes on fire.
The VIMD does not record specific details of the objects that caused the burns in the 14.3% of burns and scald hospitalisations that were attributed to hot objects. There were no cases of burns by hot objects in the VISS admissions data.
VISS data on emergency department non-admissions and the GP data indicate that women were mostly burnt when handling hot cookware (pots, pans, dishes and trays), touching hotplates, stoves and ovens when cooking or, to a lesser extent, when they fell onto a heater or fell asleep in front of a heater.

Recommendations

Strategies and countermeasures

Promote the availability and widespread installation of single purpose lithium smoke alarm batteries (which last ten years) and electric safety switches (power outlet, permanently installed switchboard units and portable units) in existing homes.
The MFB and the CFA should trial and evaluate the effectiveness of a local doorstop purchase/free installation smoke alarm scheme scheme in association with a local area survey on smoke alarm compliance in at leased one defined area.
Adapt and extend the current multidimensional Victorian child scalds prevention campaign, Hot water burns like fire to cover community living older people, and adults with disabilities, conditions and illnesses (particularly epilepsy) that put them at higher risk of burn and scald injuries.
Implement outreach burn and scald prevention programs (which include installation of smoke detectors) for homebound older people and people with disabilities (see safety guidelines below).
Implement general community education, to raise awareness among women of the hazards posed by boiling water, hot oil and steam and to re-enforce the cooking and handling techniques which reduce the risk of burns and scalds.
Advocate the development and implementation of an Australian & New Zealand Standards that require cigarettes to be fire-safe (self-extinguishing) in collaboration with international developments.
Upgrade fire and burn safety education in secondary school level courses, particularly in Home Economics, and develop or make available resources to assist students to identify burn hazards and solutions when designing their individual practical projects at the more senior levels of study.
Identify industry or sponsorship support for the further development of a spill-resistant mug to reduce scalds from hot drink spills.
Approach manufacturers of stoves and cooktops to seek design changes to reduce the risk of ‘granny gown’ and contact burns and investigate the incorporation of US developments to prevent cooker-top fires. .
Develop and trial protective sleeves made of fire-resistant material with elasticised wrists to prevent clothing ignitions while cooking (to be promoted particularly to older people and to people with disabilities affecting their manual dexterity and reach).
Encourage food manufacturers and distributors to place warning labels (including instructions on safe frying techniques and the correct response to fat and oil fires) on cooking oils and frozen products cooked by shallow and deep-frying.
Encourage manufacturers, importers and consumer authorities to conduct a product safety review of stoves and cooktops.

Surveillance, research and investigations

Produce guidelines to improve one-line case narratives on data surveillance systems, for example CFS, so that the ignition source of the fire, premorbid conditions of victims and alcohol involvement are routinely reported.
Improve quality and accessibility of data collection systems on fire-related injuries and institute systematic monitoring and routine sharing of information among agencies with responsibility for prevention and control measures.

7. Homicide and assaultive injury

The fatal assaultive injury cases in the home are mostly the result of incidents of domestic violence, which is variously defined as ‘partner’, or ‘spousal (legal or defacto)’ violence or violence between ‘intimates’. Violence perpetrated by family members is sometimes included in the definition of domestic violence. In the analyses of CFS and VISS case narrative data boyfriends and ex-boyfriends were grouped with other ‘partners’, and violence perpetrated by other family members was grouped separately. The lack of an agreed definition of ‘domestic violence’ complicates the estimation of incidence in an area where data is necessarily drawn from a number of sources (e.g. health, police and crime statistics and community surveys).

Homicide and assaultive injury is the third-highest ranked cause of female injury death in the home, accounting for 7.9% of home injury deaths.
Assaultive injury is a low- to middle- ranked cause of VIMD hospital admissions and VISS emergency department presentations, accounting for 1.7% and 5.1% of home injuries respectively.
According to ELVIS GP data, assaultive injury is a low-ranked cause of GP presentations, contributing 1.4% of home injuries. This is probably an underestimation because the ELVIS data are drawn from one rural region.

It was difficult to accurately estimate the proportion of domestic and family violence cases on the CFS and hospital-based databases because of inconsistent reporting in the one-line case narratives of information on the perpetrators of the assaults (including relationship to the victim). Under reporting is a well-documented feature of all data collections on domestic and family violence.

Deaths (fatal assaults)

Sixty-three adult female fatal home assaults were recorded on the CFS database over the 5-year period 1989-90 to 1993-94 (an average of 12-13 fatal assaults in the home per year). They are mainly shootings and stabbing.
A substantial proportion of these fatal assaults are the result of ‘domestics’ but an accurate estimation of how many is not possible because the assailant’s relationship with the victim was only specified in just over one-half (54.0%) of the fatal assaults recorded on the CFS.
In approximately 85.0% of cases where the assailant was reported as known to the victim, the assailant was identified as a partner (‘intimate’) of the victim ie. the former or current husband, de facto, ‘partner’ or boyfriend. In a further 6.0% of cases the assailant was a family member or related by marriage to the victim. In only one narrative was the assailant described as unknown to the victim.

Hospital admissions

The poor reporting of the place of occurrence (location) of injury in VIMD hospital admissions database prevents an accurate estimation of the incidence of hospitalisations from violent episodes in the home.

There were 399 cases of assaultive home injury hospitalisations recorded on VIMD from 1987/88 to 1993/94 (annual average 57 cases). However, only about one-fifth of all female assaultive injury cases on VIMD in that period (n= 2,633) were coded for place of occurrence of injury. Approximately 71.0% of the coded cases were for assaultive injury that occurred in the home. If the incidence is re-calculated on this basis it is likely that the annual number of hospitalisations from assaults in the home is in the vicinity of 270 cases. The accuracy of this estimation depends on whether the group of cases coded for location is representative of all cases of assaultive injury (and this could not be checked). This estimate does not take into account undetected cases of domestic violence that are admitted to hospitals.

Hospital emergency department presentations (including admissions)

Assaultive injuries are a mid-ranked cause of female home injury presentations to VISS hospital emergency departments, accounting for 5.6% of home injury presentations.

Unarmed fights (65.3%), assault by cutting and piercing (7.1%) and assault by blunt or thrown objects (10.3%) are the significant causes of injury.
The assailant is most frequently a current or former partner (husband, defacto or boyfriend) of the victim (representing 56.1% of cases where information on the assailants’ relationship to the victim was given in the case narratives).
A wide range of ‘first degree’ relatives (mostly sons and brothers) and relatives by marriage inflicted the assaultive injuries by family members other than partners (representing 16.3% of cases where information on the assailants’ relationship to the victim was given in the case narratives).

General Practitioner presentations

Only a small proportion (1.4%) of ELVIS GP presentations were for assaultive injury. Compared to other reports this is a comparatively low figure. It may be explained, in part, by the rural nature of the ELVIS collection (which may result in under-reporting) and the unavailability of GPs in the evenings and at weekends in rural areas which is the peak time for assaultive injury presentations to emergency departments (as shown in VISS data). Also, for reasons of privacy, in rural areas women may choose to attend an emergency department with assaultive injury rather than present to their GPs.

Methods of assault

The majority of homicides involved a weapon. Guns (31.7%, mostly shotguns) and cutting and piercing instruments (30.2%, presumably knives) were the most frequently used weapons.
At the other levels of severity, injurious assaults are mostly unarmed hitting incidents. VISS emergency data suggest that a substantial proportion of these occur during domestic arguments.

Age of injured women

At all levels of injury severity women aged less than 40 years appear more at risk of assaultive injury than do older women. There was also a peak of homicides in women aged 65-74 years but this pattern was not evident at the other levels of severity. This could be a manifestation of under-detection. Older women may be less likely to report domestic assault as the cause of their injuries or may be less likely to be identified as victims of assault by hospital emergency department staff and general practitioners.

Time of injury event

VISS data on emergency department presentations shows that there is a higher frequency of assaultive injury presentations at weekends, starting from Friday at around 8pm, peaking on Saturday and Sunday nights from 7pm. There were also peaks in the evenings of weekdays. This pattern has implications for the provision of referral services.

Recommendations

The tentative nature of these findings highlight the problems with available data on assaultive injury in the home, particularly in relation to domestic violence, and the need for systems-wide improvements in data collection on domestic violence.

Strategies and countermeasures

Train and support hospital emergency staff and GPs to identify, record and deal with domestic violence cases.
Ensure that domestic violence referral services operate at night and on weekends.

Surveillance, research and investigations

Improve the compatibility of existing data systems that record cases of domestic and family violence
Improve data collection and recording of domestic violence on CFS, VIMD and emergency department injury surveillance systems:

ensure that the assailant’s relationship with the victim and other useful information on the circumstances of the injury event are systematically reported in CFS and VEMD case narratives;

improve the reporting of the place of occurrence (location) of injury in VIMD hospital admissions database and the new emergency department system (VEMD);

introduce a locally expanded ICD9 E-code classification to identify the perpetrator and record other circumstances of injury in cases of interpersonal injury and abuse presenting to hospitals; and

encourage GPs to systematically record cases of domestic violence.

The lower-ranked causes of home injury —natural and environmental injury; choking, suffocation and foreign bodies in orifice; and overexertion and strenuous exercise— are dealt with in the body of the report (chapters 15-17).

Sponsor: Department of Human Services, Victoria