Building Haiti back better: health sector lessons from the 2004 Indian Ocean tsunami
by Richard Garfield and Erin Chu March 2010

As a humanitarian crisis, the Indian Ocean tsunami in 2004 is comparable to the 2010 earthquake in Haiti. In Aceh in Indonesia, the most affected area, 230,000 people died and half a million were made homeless. By comparison in Haiti, roughly 200,000 people died and a million were made homeless. Indonesia had competent central government, but reconstruction in affected areas in Aceh was made more difficult by ongoing insurgencies there. While Haiti had no insurgency at the time of the earthquake, it suffered from weak governance, a near-total lack of governmental services and the inheritance of repeated insurgencies and dictatorships. After five years, recovery programmes for the tsunami recently ended, giving us many lessons towards planning for recovery in Haiti.

A lot of the ‘right’ rhetoric is being heard in Haiti: that reconstruction must build back better and smarter, and that it must be led by Haitians. But these challenges may be greater in Haiti than they were in Aceh. There is less capacity overall in the Haitian state, fewer training institutions are available and there is less routine income with which the state can work. There are more NGOs (270 registered with the health cluster alone, as of 3 February – perhaps three times more in all), used to working with little coordination and expecting even less from the government. More Haitian doctors and nurses live outside the country than within it, and emigration will continue to draw away the skilled staff necessary to create a stable society and adequate health care system.

 Recovery contributions

Aceh, with a population of four million, received more than $5 billion for reconstruction. Haiti – population nine million – has received initial reconstruction pledges of $1bn. Prior to the earthquake, extreme poverty in Haiti was greater than in Aceh, yet malnutrition was not. Global acute malnutrition in Haiti was around 5%, while in Aceh it was between 8% and 16%. Haitians used a variety of subsistence strategies, including pooling resources among families, using charity services and remittances (under normal conditions, a third of all Haitian income comes from remittances from family members outside the country).[1] These coping methods can be leveraged in reconstruction, using large-scale investment in agriculture to produce more food, and to engage community health workers to feed and monitor children. Recovery for Haiti must also attend to the needs of the country as a whole, rather than the capital alone. Five years on from the tsunami, the rural poor remain in Aceh remain at a health disadvantage.

 Health data and monitoring

No reliable region-wide surveys were available for Aceh prior to the tsunami. By contrast in Haiti a national Demographic Health Survey (DHS) in 2008 provides good baseline data. An initial rapid assessment (IRA) was carried out within three weeks of the earthquake, and plans are being made for periodic recovery monitoring.

A DHS carried out in Aceh in 2008 showed that the prevalence of health problems among children remained higher in the province compared to the country overall, demonstrating continued problems in coverage and effectiveness several years after the tsunami. Over the last decade Haiti greatly reduced malaria, HIV and infant mortality levels. Still, overall mortality is more than 50% higher than in Nicaragua, the next-poorest Latin American country, or the neighbouring Dominican Republic. Haiti’s injury-related deaths prior to the earthquake were nearly double those of these other two countries, while communicable, perinatal, maternal and nutrition-related diseases were three to five times more common.

Malnutrition, malaria and immunisation-preventable diseases in most tsunami-affected countries have declined, but the overall change is modest. Areas with social indicators that were already improving rapidly have returned to rapid improvement, while areas improving more slowly before the tsunami have recovered only enough to resume that slow improvement. Five years after the tsunami, in all areas, the major limitations to improvement in health are the same as they were before. While some agencies had personnel on the ground for years prior to the tsunami and understood these local conditions, that information was seldom applied to tsunami recovery programming. Indeed, much of the programming in years two to five looked more like extensions of programmes in the first emergency period, rather than transformative recovery programmes. Thus, health conditions today overall are much as they were prior to the tsunami – a far cry from the transformed system large-scale international funding might have provided.


Mental health

Health services developed in tsunami recovery provided well for the emergent physical needs of families, but were generally less prepared to address social and psychological needs, develop health policy or improve the supervision and productivity of health workers. Few programmes focused on the psycho-social needs of survivors, and those that did (mainly short-term training programmes) were not combined with major programmatic initiatives and have largely been forgotten. The tsunami also reinforced the importance of using existing resources to establish new services. For example, mental health programmes could be provided in primary care clinics, which are easily accessible and are less likely to stigmatise patients requiring mental health treatment.[2] Not surprisingly, involving beneficiaries in the aid process helped them regain control of their lives and led to improvements in well-being.



Training and systems development was almost entirely missing from the recovery programme in Aceh. Building on the enhanced investments in facilities and the training of primary level health care workers, this should have included training district and regional health system managers, continuing-education officers and epidemiologic analysts. Such higher-level training would have developed systems capacity to coordinate new health resources more effectively.

New maternal health workers were trained to take the place of those lost in the tsunami. However, when these new nurses and midwives began to graduate in large numbers in 2007 they had limited skills and experience; their training did not give them the confidence to perform key services, and was not tailored to the epidemiological conditions of the country. Midwifery graduates from cities were less willing to serve in the remote areas where they were needed, nurses left affected regions with newly salable skills and foreign doctors were sometimes employed at unsustainably high cost. Policies to train people from remote areas, to create systems of career advancement and to train and equip new health workers to address major local threats to health could have built back smarter, and better.


Health infrastructure

Donors in Aceh supported the building and equipping of a large network of primary care centres, including many focusing on child welfare and nutrition and the use of midwives as health promoters. Prior to the earthquake, there were similar plans in Haiti to strengthen the role of community health workers in villages and focus more strongly on a coordinated network of primary care facilities. Compared to Aceh, Haiti lost fewer doctors and midwives but more nurses in the earthquake. It lost far fewer health facilities, but many of these were the major institutions in the country, including the country’s 11 largest hospitals. Without strong direction from the Haitian Ministry of Health, health reconstruction efforts are likely to be uncoordinated, yielding very unequal care, focused on diseases of lesser epidemiologic importance and preoccupied with running and maintaining hospitals to the detriment of preventive and primary health care.

After just three years of reconstruction, Aceh reached near-sufficient numbers of health workers and replaced 517 destroyed health facilities with 1,115 new structures.[3] Thus, many agencies literally built back better after the tsunami in the construction of health facilities, but were less successful at developing the capacity needed to use these improved facilities. International agencies and the government in Aceh in particular focused much of its attention on building new health facilities when greater efforts could have been made to develop capacity to organise and run health and nutrition programmes dealing with the region’s major needs in nutrition, diarrhea and respiratory disease.

This happened, in part, because large-scale destruction of infrastructure forced humanitarian agencies to become building contractors. Building services could instead have been contracted out to companies with expertise in construction. At the same time, centres could have started with temporary one-room or rented facilities, growing depending on patterns of use and construction capacity. It is important in reconstruction that health programmes should not be entirely dependent upon physical structures.

Epidemics and disease outbreak

While everyone was looking for a non-existent cholera epidemic in Aceh, tetanus peaked weeks after the tsunami with 107 reported cases from hospitals and health centres, with a case fatality rate of 19%.[4] A tetanus epidemic is now underway in Haiti, and measles, malaria and dengue epidemics may occur.

There was a dramatic increase in the number of acute respiratory infections (ARI) during the first months after the tsunami. Respiratory infections, in normal circumstances, are the leading cause of morbidity and mortality in developing countries. High levels of respiratory and diarrheal diseases emerged within a week of the Haitian earthquake due to crowding in open urban areas, lack of sanitary facilities and low pre-earthquake immunisation levels.

Disease surveillance was reinitiated in Aceh by the second week of February, and an epidemic and alert response team, in collaboration with the Provincial Health Office, developed a surveillance system that included NGO, hospital and laboratory activities. A six-month activity plan for the rehabilitation of provincial and district health offices in Aceh was also established, to develop longer-term surveillance capacity.[5] A similar process is underway in Haiti.


Long-term planning

Many successful emergency actions in Aceh were not articulated with long-term development goals for the health system. Aid workers were more engaged in programmatic activity than in strengthening local planning and management capacity. In retrospect, it is clear that these actions worked to rapidly re-establish a minimal system of care. But inadequate attention to pre-existing weaknesses in the government’s health and administrative systems has produced the same key constraints that existed prior to the tsunami. More attention to developing district- and provincial-level capacity for planning and administration would have helped to further advance programmes and integrate them into normal country programmes.

Poor governance in Aceh had much to do with the inefficient use of aid. This is why engagement and investment in Haiti must be done in consultation with Haitians, and in a way that builds administrative capacity and a national consensus to contribute more to Haiti’s long-term development. Despite worsening economic conditions and unstable governance, Haitians have managed to send their children to primary school and get them immunised at ever-increasing rates since 1990. Improving the coverage and quality of these and other basic services can help transform collective despair into hope.




  • Strategic planning towards multi-year programming should replace emergency response approaches as soon as possible. Training of local staff for this needs to emphasise the articulation of facilities construction planning, training of personnel and the establishment of effective administrative systems.
  • Assessment capacity and structures must be created to monitor priority indicators for the supervision and management of the health system. This is best implemented within existing organisational structures, rather than by creating new entities. Expanded monitoring is valuable only if capacity is strengthened in existing systems for routine monitoring, instead of special stand-alone systems.
  • Agencies should not be distracted by physical building programmes, especially when construction can be contracted to more experienced sector specialists.  
  • Strengthening rural primary care with community health workers and auxiliary nurses will be the most effective means to improve health, while also providing the employment needed to help stabilise these communities.

The tsunami events catalysed efforts to implement the Cluster Approach. This is a challenge for Haiti, where small agencies operate freely and the government has been a weak partner.

  • In the transition from emergency relief to development, actors must recognise the need for multi-year strategic planning as early as possible.
  • Use aid to support communities at the onset of disaster. Giving ownership to beneficiaries will make humanitarian actors accountable to affected communities, and improve the appropriateness of interventions. Furthermore, it can build capacity to deal with future vulnerabilities, which will outlast the commitment of short-term relief.


Richard Garfield, RN DrPH, is Henrik H. Bendixen Professor of Clinical
International Nursing at Columbia University. Erin Chu, MPH MPA, is a consultant.


[1] E. Gibbons and R. Garfield, ‘The Impact of Economic Sanctions on Health and Human Rights in Haiti’, American Journal of Public Health, 1999; 89(10): 1499-1504.


[3] The Executing Agency of Rehabilitation and Reconstruction for Aceh and Nias- BRR, Book Series. Book C- Map. Spread of Efforts and Achievements, 2009.

[4] D. Guha-Sapir and W. Panhuis, The Andaman Nicobar Earthquake and Tsunami 2004: Impact on Diseases in Indonesia (Brussels: Center for Research on the Epidemiology of Disasters (CRED), 2005).

[5] Ibid.