Correspondence 2 1·00 ODA/GNI % 0·80 r=–0·68 0·60 3 0·40 0·20 0·00 0·20 0·25 0·30 0·35 0·40 Gini coefficient Figure: Correlation between official development assistance (ODA) as a percentage of gross national income (GNI) and Gini coefficient inequality effects from those of individual income, their being inextricably linked. Ecological studies have also suggested that income inequality represents a correlate of operating social policies, such as those concerning health care, education, social welfare, and working conditions.2,3 Moreover, associations have been found between income inequality and extent of social capital in terms of civil trust, links between individuals, and reciprocity.4 From an economic perspective, a nation’s income inequality sets the level of redistribution of public resources, which are critical in defining the protective role of social policies. Our analysis indicates that this wealth redistribution is also related to the altruistic behaviour of a nation—ie, communities that tend to distribute their goods equally are also keen to aid people in poorer countries. As with human love, whereby the more individuals receive the more they give back, for countries, the more they are equalitarian, the more they grant externally. Sociologists might like to explore these behaviours. We just argue that this is an additional reason to include Gini coefficient in the set of key human development indicators of international agencies. We declare that we have no conflict of interest. *Enrico Materia, Lorenza Rossi, Gabriella Guasticchi materia@asplazio.it Agenzia di Sanità Pubblica, Regione Lazio, Rome, Italy 1 Diez-Roux AV, Link BG, Northridge ME. A multilevel analysis of income inequality and cardiovascular disease risk factors. Soc Sci Med 2000; 50: 673–87. www.thelancet.com Vol 365 April 23, 2005 4 Kaplan GA, Pamuk ER, Lynch JW, Cohen RD, Balfour JL. Inequality in income and mortality in the United States: analysis of mortality and potential pathways. BMJ 1996; 312: 999–1003. Lynch JW, Davey Smith G, Hillemeier M, Shaw M, Raghunathan T, Kaplan GA. Income inequality, the psychosocial environment, and health: comparisons of wealthy nations. Lancet 2001; 358: 194–200. Kawachi I, Kennedy BP, Lochner K, Prothrow-Stith D. Social capital, income inequality, and mortality. Am J Public Health 1997; 87: 1491–98. Seeking a safer blood supply In her World Report (Feb 12, p 559),1 Barbara Fraser includes comments that I made on some of the factors that affect blood safety in our region, and mentions the initiative led by the Pan American Health Organization (PAHO) to improve it. Issues regarding the availability and safety of blood still remain unresolved in a good number of American countries.2,3 However, progress has been made as a result of a programme supported by the Gates Foundation (table), which was designed to improve the safety of blood for transfusion by screening 100% of blood units for infectious markers, implementing programmes of external assessment of performance in all blood banks, and collecting blood from voluntary blood donors. From 2000, before the Regional Initiative was launched, to 2003, the last year for which we have data from the countries, the estimated risk of contracting a viral infection through transfusion diminished from 1 in 4011 donations to 1 in 41 858 in the all Latin American and Caribbean countries; the risk of Trypanosoma cruzi infection in continental Latin America decreased from 1 in 762 to 1 in 3340 in the same period. The figures for 2003 are still unacceptably high owing to a high prevalence of markers among donors and the lack of universal screening. We have estimated that at least 108 217 viral infections were prevented in the 3-year period by laboratory screening, whereas 481 infections were transmitted because of lack of screening. The major obstacle for universal screening is the emergency collection of blood because of low stock in the blood bank. The current blood systems, well described by Fraser,1 generally do not promote voluntary blood donation and, as is common in Peru, make the patients’ relatives responsible for recruiting potential donors. Payments are often made to donors by the patients’ relatives, who also must purchase the blood collecting bag and pay for the laboratory tests. This approach has been implemented because the hospitals do not have resources to invest either in the promotion of voluntary blood donation or in the supplies. The Director of PAHO opined that the national strategies “should be geared toward taking blood donation out of the hospital environment”, and that “the ministries of health. . . assume a leading role in guiding all stakeholders and in bringing together the resources that are needed to manage the blood for transfusion as a valuable national resource”.4 Unless these goals are pursued, availability and safety of blood for transfusion in Latin America and the Caribbean will remain unsatisfactory. Improvements made in recent years seem to have reached a ceiling that will not be overcome unless major changes are implemented by the national health authorities. PAHO has developed a plan of action and strategies to improve the safety of blood Number of blood units collected Donation rate per 1000 inhabitants Proportion of units screened for: HIV Hepatitis B virus Hepatitis C virus Treponema pallidum Trypanosoma cruzi Total number of blood banks Number doing external assessment of performance Voluntary blood donations Risk of viral infected-transfusion Risk of T cruzi-infected transfusion 2000 2003 6 409 596 12·7 7 325 093 13·9 99·66% 99·65% 98·79% 99·57% 78·98% 4738 1137 (24%) 99·93% 99·86% 99·52% 99·84% 88·09% 2509 1330 (53%) 15·4% 1 in 4011 1 in 762 36·1% 1 in 41 858 1 in 3340 Table: Change in indicators of safety of blood in Latin American and Caribbean countries, 2002–03 1463 Correspondence transfusion—not just the transfused products—in the next 5 years. This plan is expected to be implemented as a Regional Program. I declare that I have no conflict of interest. José Ramiro Cruz cruzjose@paho.org Regional Advisor Laboratory and Blood Services, Pan American Health Organization, 525 23rd Street NW, Washington, DC 20037, USA 1 2 3 4 Fraser B. Seeking a safer blood supply. Lancet 2005; 365: 559–60. Cruz JR, Perez-Rosales MD. Availability, safety and quality of blood for transfusion in the Americas. Pan Am J Public Health 2003; 13: 103–09. Schmunis GA, Cruz JR. Safety of the blood supply in Latin America. Clin Rev Microbiol 2005; 18: 12–29. Roses Periago M. Promoting quality blood services in the Region of the Americas. Pan Am J Public Health 2003; 13: 73–74. Barbara Fraser’s World Report1 on blood transfusion safety highlights the critical shortcomings of transfusion practices in the developing world. More than 80% of the world’s population has access to only 20% of the world’s safe blood supply.2 And it is this population that has a high demand for blood products. Southeast Asia, for example, collects only 7 million units of the 15 million units of blood that it requires annually. Many of the 150 000 women who die each year from blood loss due to pregnancy-associated causes might live if sufficient blood is available.3 This gap between supply and demand further compromises the adequate provision of safe blood. In such situations, blood is more likely to be bought from private blood banks, which are more inclined to compromise safety standards. The global contribution of developing countries to blood collection from altruistic donors stands at a meagre 16%.3 Replacement and paid donors provide the bulk of donations. Indeed, unsafe blood accounts for 8–16 million hepatitis B virus infections, 2·3–4·7 million hepatitis C virus infections, and 80 000–160 000 HIV infections annually.3 Aside from the shortage problem and transfusion-transmitted infections, the complications associated 1464 with transfusion reactions in these settings are largely unknown. Pakistan is a country with a human development index of 142. The reported prevalences of hepatitis B and C are 5–8% and 7–10%, respectively. About 50% of blood banks in Karachi, the largest city in Pakistan, regularly employ paid blood donors.4 An average patient receiving two units of blood has a 10% chance of acquiring hepatitis C infection.4 Altruistic donors account for only 5% of all donors. The national blood policy announced in 2003 has been termed a step in the right direction. As a result, more than 90% of the blood is properly screened in public hospitals. However, there are shortcomings in implementation, particularly relating to the private blood banks which greatly outnumber public blood banks. It is important to remember that poverty remains the underlying cause in the developing world for the present situation. Weak infrastructure, illequipped centres, poorly trained staff, inadequate policy implementation, and frequent power breakdowns are problems needing rectification for a sustainable solution. Although badly needed, altruistic donors in developing countries will be difficult to come by owing to wellestablished traditional beliefs and practices.5 The perceptions and concerns of people regarding transfusions must be assessed to formulate targeted educational strategies and mobilise donors. It is also important to enforce transfusion guidelines strictly to avoid unnecessary transfusions. Donor screening is an effective although underused strategy in the developing countries. Questions about high-risk behaviour are seldom asked.4,5 The high cost of screening assays often leads to blood being screened only if the patient is willing to pay.4 Screening costs must be reduced by developing cost-effective assays, pooled testing,5 and eliminating import duties on essential equipment. A concerted effort to address these problems will significantly improve the situation in the years to come. We declare that we have no conflict of interest. *Fawad Aslam, Junaid Ali Syed fawadaslam2@hotmail.com Male Hostel, Aga Khan University Medical College, Stadium Road, Karachi 74800, Pakistan 1 2 3 4 5 Fraser B. Seeking a safer blood supply. Lancet 2005; 365: 559–60. Larkin M. WHO’s blood-safety initiative: a vain effort? Lancet 2000; 355: 1245. World Health Organization. Safe blood starts with me! Blood saves lives! Stories and souvenirs from World Health Day 2000 together with useful information on blood safety. http://www.who.int/bloodsafety/en/ WHD_Safe_Blood_2000.pdf (accessed Feb 21, 2005). Luby S, Khanani R, Zia M, et al. Evaluation of blood bank practices in Karachi, Pakistan, and the government’s response. Health Policy Plan 2000; 15: 217–22. Wake DJ, Cutting WA. Blood transfusion in developing countries: problems, priorities and practicalities. Trop Doct 1998; 28: 4–8. CT colonography for detection of colon polyps and cancer The mediocre results obtained for computed tomographic colonography (CTC) by D C Rockey and colleagues (Jan 22, p 305)1 raise questions about the examination techniques used in conducting this important new imaging procedure. Rockey and colleagues do not provide key data in several areas. The first is colon preparation. What measure was used to assess adequacy of gas insufflation? This is an essential part of “preparation”, but was not directly addressed in the report. Also, was there any attempt to score retained fluid which is commonly found with preparation regimens used for colonoscopy? Polyps are sometimes hidden by dependent fluid collections. Were there explicit formal criteria by which a site or reader could initially participate in the study and then be retained subsequently on the basis of number of patients contributed and performance? Are there any readerspecific or site-specific performance data to further illuminate the apparent lack of effect of previous experience? Moreover, the radiologists were not www.thelancet.com Vol 365 April 23, 2005