The world health report

Chapter 4


Other risks to health

Clearly, many thousands of other threats to health exist within and outside the categories outlined above. These include very large causes of disease burden, such as risk factors for tuberculosis (see Box 4.6) and malaria (which is currently responsible for 1.4% of global disease burden, with the vast majority of burden from this disease among children in sub-Saharan Africa). Genetics plays a substantial role in attributable burden (see Box 4.7). Technological developments could lead to considerable avoidable burden. In general, the approaches and methodology outlined in this report can be applied more widely, and as a result the potential for prevention by focusing on causes of disease can be further refined. Two other groups of risk factors are described below (see Table 4.8).

Box 4.7 Genetics and attributable and avoidable burden

It is a common misconception that diseases are caused by either genetic or environmental factors; almost all diseases are caused by both. Although it is not possible to estimate the attributable burden of disease from "genetic causes", it is potentially possible to estimate the burden attributable to certain gene mutations or alleles.

Diseases caused by mutations in single genes, such as phenylketonuria, tend to be rare, whereas the genetic influences on common causes of morbidity and mortality are more complex. In some cases single gene mutations which carry a high risk of disease can be identified but do not necessarily have a major impact on the incidence of disease in populations. For example, gene mutations which confer a high risk of breast cancer are important for carriers of those mutations but are present in only a small proportion of women who develop breast cancer.

Recent developments in genetics offer substantial potential for health gain through increasing the understanding of the biological basis of diseases, identification of high-risk individuals enabling targeted risk factor modification, and the potential for tailored treatment. The greatest possible gains lie in more direct applications. Pharmacogenetics promises to allow drug prescribing to be tailored to individuals likely to have most benefit or least susceptibility to adverse drug reaction. More important yet may be the discovery of disease susceptibility genes that allow identification of a protein in which altered function affects the disease process. This in turn could lead to interventions. While the avoidable burden of genetic disease cannot yet be quantified, especially for common chronic diseases that are influenced by multiple genes, it is likely to be substantial even if only a small fraction of the attributable burden is reversed.

The coming decades will see improved prevention and treatment through appropriate mixes of new genetic and traditional preventive strategies. Nonetheless, ambitious targets need not await these new interventions. Combinations of primary prevention, focusing on major risk factors, and secondary prevention have already achieved substantial reductions in major chronic diseases in just a few decades, during which time gene pools did not essentially alter. For example, age-specific reductions of 25--75% have been achieved in breast cancer mortality in the United Kingdom and United States, coronary disease in the United States and Scandinavia, stroke in Japan, and lung cancer in the United Kingdom. The potential to repeat such successes will clearly be greater if preventive efforts can be augmented by appropriate genetic-based interventions.

Sources: (93,98).

Box 4.6 Risk factors for tuberculosis

About 9 million new cases of tuberculosis (TB) occur each year. Including people who are also infected with HIV/AIDS, approximately 2 million patients die from TB annually. The global caseload is almost certainly rising, driven upwards in sub-Saharan Africa by the spread of HIV/AIDS and in Eastern Europe by the deterioration of health in general and of TB control in particular. There is a large reservoir of cases in Asia, and TB remains one of the most significant causes of ill-health and premature mortality.

One of the reasons for the persistent burden of tuberculosis is a failure to address the principal risk factors. The risks associated with TB can be put in three groups: the process of infection, progression to disease, and the outcome of a disease episode. Environmental factors that govern exposure to infecting bacilli include crowding, hospitalization, imprisonment, ventilation and the ambient prevalence of infectious (mostly sputum smear-positive) disease. Among factors that influence the progression to disease following infection, HIV co-infection is outstandingly important; others are age, sex, diabetes, tobacco, alcohol, TB strain virulence, and malnutrition. Factors that affect the outcome of a disease episode include where treatment is given (e.g. public or private sector), whether treatment is interrupted, and drug resistance. The adverse outcomes most commonly measured are treatment failure and death. Some other risk factors for TB are commonly invoked but ill defined, ethnicity and poverty among them. Ethnicity is often a marker for specific disadvantages, such as restricted access to health services.

While the study of risk factors is a necessary part of planning for TB control, it is not sufficient. Some major risk factors may not be amenable to change, at least as they are currently defined: there is nothing to be done about age per se, though one could investigate why, physiologically, adults are at greater risk of progressing to active disease than children. Further, the risk factor approach (based on observed variation) cannot be used to examine potentially effective interventions that do not yet exist. The absence of a new vaccine is not usually thought of as a risk factor for TB and yet common sense, backed by mathematical modelling, shows how effective immunization could be.

Despite some promising laboratory research, there is unlikely to be a new TB vaccine or drug before 2010. Meanwhile, the principal question for operational research is how to strengthen present curative services. With only 27% of new infectious cases being enrolled in DOTS therapeutic programmes, the main goal of TB control is to ensure broad national coverage rather than to target specific groups at risk. In this respect, it is important for patients to recognize the symptoms and know where to seek help, to receive the correct diagnosis and drug regimen, and to understand the importance of completing a course of treatment. There are some challenging questions here, whether or not they are framed in terms of risk factors: for a social intervention like DOTS, careful thought must be given to the design of case-control studies or randomized controlled trials, and still greater caution is needed when generalizing from the results.

Sources: (91,92).

Unsafe health care practices

As well as their substantial benefits, health care practices may be a source of disease and death. In developing countries, nosocomial infections are increasingly recognized as a major problem in health care quality, although the burden of disease is difficult to estimate. Poor injection practices, including injection overuse and unsafe injection practices, constitute a subset that can be addressed because it is ubiquitous, has been studied in many countries and is associated with a particularly high toll of infection with bloodborne pathogens. Epidemiological studies have reported an association between injections and infection with bloodborne pathogens, including hepatitis B virus (HBV), hepatitis C virus (HCV) and human immunodeficiency virus (HIV) (99,102). The causal nature of this association is supported by many criteria.

A safe injection is one that does not harm the recipient, the provider or the community. In reality, many injections in the world are unsafe. The risk to the community through unsafe sharps waste disposal has not been assessed, but is probably low. The risk to the provider (i.e. needlestick injuries, see Box 4.4) was studied among other occupational risks. The risk to the recipient is mainly secondary to the reuse of injection equipment.

Because injections are overused in many countries, unsafe injections have caused a substantial proportion of infection with bloodborne pathogens, accounting for an estimated 30% of hepatitis B virus infection, 31% of hepatitis C virus infection, 28% of liver cancer, 24% of cirrhosis and 5% of HIV infections. Overall, about 500 000 deaths (0.9%) are attributable to unsafe injection practices in medical settings worldwide, the attributable fractions are highest in South-East Asia, WPR-B and EMR-D. This results in about 10.5 million DALYs (0.7%), with 39% of this burden occurring in SEAR-D and 27% in WPR-B. In these areas, unsafe injections result in about 0.7--1.5% of all disease burden. These estimates are based upon a mathematical model that was validated by epidemiological studies in most regions in the case of HBV and HCV infection. In the case of HIV infection, there is more uncertainty about the region-specific estimates, due to a lack of epidemiological studies. However, studies have been conducted in sub-Saharan Africa, where most HIV infection occurs, providing more confidence in the overall magnitude of attributable burden, and pointing to the importance of this particular mode of HIV transmission.

Unsafe injections are one form of risk in medical settings; some of the other risks are illustrated in Box 4.8.

Box 4.8 Risks in the health care system

The complex combination of processes, technologies and human interactions that constitutes the modern health care delivery system not only brings significant benefits, but also an inevitable risk in the form of adverse events. This derives from the inherent risk of measurable harm in practice (human shortcomings), products (substandard or faulty products, side-effects of drugs or drug combinations, and hazards posed by medical devices), and procedures and systems (the possibility of failures at every point in the process of care giving). These risks are associated with different health care settings -- hospitals, physicians' offices, nursing homes, pharmacies, and patients' homes.

Studies estimate the probability of patients suffering measurable harm in acute care hospitals at an alarming 16.6% in Australia, 3.8% in the United States, and around 10% in Denmark, the United Kingdom and a number of other European countries. Adverse events exact a high toll in disability and death, as well as in financial loss. Medical errors cause several tens of thousands of deaths annually in the United States alone. Although some deaths occur among people at high risk of death from their initial conditions, the loss of life years is still likely to be substantial. Estimates from the United Kingdom place the cost of additional hospital stays resulting from adverse events at approximately US$ 3 billion a year. The erosion of trust, confidence and satisfaction among the public and health care providers must be added to these costs.

The situation in developing and transitional countries is not well known, but could be worse than that in industrialized nations because of counterfeit and substandard drugs and inappropriate or poor equipment and infrastructure.

The systems view is that risk is shaped and provoked by "upstream" systemic factors that include an organization's strategy, its culture, its approach towards quality management and risk prevention, and its capacity for learning from failures. System change as a means to reduce risk is therefore more potentially effective than targeting individual practices or products.

Sources: (103,115).

Abuse and violence

Abuse and violence are major causes of disease burden worldwide and there are many types: violence between individuals, including intimate partner violence, and collective violence orchestrated as part of wars and genocide. These are further outlined in Box 4.9. Child sexual abuse is another major component of burden resulting from abuse and violence in society.

Box 4.9 Violence

In 2000, violence caused 700 000 deaths in the world: about 50% by suicide, 30% by interpersonal violence, and 20% by collective violence.

Interpersonal violence

Interpersonal violence is defined as "the intentional use of physical force or power, threatened or actual, against another person that results in or has a high likelihood of resulting in injury, death, psychological harm, `maldevelopment' or deprivation". As well as violence by strangers and acquaintances, it includes child maltreatment, spouse abuse, elder abuse and sexual violence. The true number of deaths is probably underestimated.

Worldwide, adolescents and young adults are the primary victims and perpetrators: interpersonal violence was the sixth leading cause of death among people aged 15--44 years in 2000. The highest estimated regional homicide rates per 100 000 population occurred in Africa (22.2) and the Americas (19.2), compared with Europe (8.4), the Eastern Mediterranean (7.1), South-East Asia (5.8) and the Western Pacific (3.4).

Many more people survive acts of interpersonal violence than die from them. Around 40 million children are maltreated each year. Rape and domestic violence account for 5% to 16% of healthy years of life lost by women of reproductive age. Between 10% and 50% of women experience physical violence at the hands of an intimate partner during their lifetime. Beyond the deaths and injuries, there are many profound health and psychological implications for victims, perpetrators and witnesses of interpersonal violence.

For individuals, risk factors include being a victim of child abuse and neglect, substance abuse, and being young and male. In families, marital discord, parental conflict, and low household socioeconomic status are important risks. In the community, low social capital and high crime levels contribute. In society generally, rapid social change, poverty and economic inequality, poor rule of law and high corruption, sex inequalities, high firearm availability, and collective violence are risk factors. In combination, these factors underlie the close relationship that exists between indicators of interpersonal violence and the socioeconomic context. Correlational studies show higher homicide rates among countries with lower per capita GDP. Findings consistently demonstrate that high levels of inequality coincide with high homicide rates and high rates of non-fatal violence among the poorest sectors of the population;

Interpersonal violence can be prevented and its destructive consequences lessened by focusing on these risk factors, ideally in combination and at different levels simultaneously. Home visits by nurses have shown effectiveness, as have various programmes on parent training, improving urban physical and socioeconomic structure, increasing protective knowledge in schools about sexual abuse, targeting the interaction between firearms and alcohol, and multimedia interventions aimed at reducing the social acceptability of violence. Almost all evaluations of such programmes have been conducted in industrialized countries. In the developing world it is projected that the burden of disease resulting from interpersonal violence will nearly double by 2020 unless preventive action is taken.

Collective violence

Collective violence is a broader term than war or conflict. It encompasses events such as genocide and applies when one group makes instrumental use of violence against another to achieve an objective. It is associated with major threats to health in what tend to be the world's poorer countries. In 2000, an estimated 310000 deaths resulted directly from collective violence -- mostly in Africa and South-East Asia.

Although a prominent feature of human history, collective violence has not received much systematic study. Today it is often characterized by varying degrees of state collapse or dysfunctional governance and a multiplicity of armed actors, often including child soldiers. Economic motivations or ethnic divisions have become more prominent causes of violence than political ideology. The results have often been indiscriminate attacks on civilians and degradation of social capital. Sometimes health infrastructure is specifically targeted, damaging access to water supplies and basic sanitation, and jeopardizing delivery of health interventions such as disease eradication programmes.

Indirect effects of collective violence arise from infectious disease, malnutrition, population displacement, psychosocial sequelae, and exacerbation of chronic disease. Mortality rates 80-fold higher than the baseline have been recorded in populations fleeing collective violence in Rwanda.

Risk factors for collective violence include the generalized availability of small arms, inequalities in access to educational, economic and political opportunities, and abuse of human rights. There is a need to combine efforts of the public health and social science sectors to guide progress in this area and to identify priority areas for intervention.

Sources: (116,117).

Child sexual abuse (CSA) encompasses a range of sexual behaviours perpetrated by adults upon children. Abuse can be non-contact (including behaviours such as unwanted and inappropriate sexual solicitation or indecent exposure), contact (such as sexualized kissing, hugging, touching or fondling) or intercourse (including any penetrative act such as oral, anal or vaginal intercourse or attempted intercourse).

The prevalence of CSA is estimated from retrospective report and is higher than many find comfortable or plausible. In the review carried out as the basis for this report, prevalence estimates were available from 39 countries in 12 of the 14 country groupings, although data quality varied considerably between countries. After controlling for differences between studies, the prevalence of non-contact, contact and intercourse types of CSA in females was about 6%, 11% and 4%, respectively. In males it was about 2% for all categories. Thus over 800 million people worldwide may have experienced CSA, with over 500 million having experienced contact or intercourse types of abuse.

Not only is CSA common, it is also damaging. Research conducted in economically industrialized countries has shown that CSA increases the risk of a range mental disorders in later life, including depression, panic disorder, alcohol and drug abuse and dependence, post-traumatic stress disorder and suicide. Risks increase with the intrusiveness of the abuse. Uncertainty remains because of the lack of knowledge about the impact of cultural differences on CSA prevalence and its relationship with mental disorders. It is, however, certain that CSA causes a considerable burden of disease. It is estimated that about 33% of post-traumatic stress disorder in females and 21% in males is attributable to CSA. The attributable fraction for panic disorders is 11% worldwide, and CSA is estimated to cause about 5--8% of self-inflicted injuries, unipolar depression, and alcohol and drug use disorders. Overall, 0.1% of deaths worldwide (79 000) are attributable to CSA. Much of the burden is disabling rather than fatal, and occurs in the young. Thus CSA causes 8.2 million DALYS (0.6%); 0.4% in males and 0.8% in females. The highest proportion of burden (1--1.5% of total) occurs in females in AMR-A, SEAR-D, WPR-A and WPR-B.