Category Archives: Health system

Operational research

Posted on by — No Comments ↓

ResearchBlogging.orgIn global health, operational research is an idea increasingly used by donors and policy makers. It involves analytical methods to help improve public health interventions and treatment of diseases in real-life situations. It is thus different from randomized clinical trials that determines efficacy of an intervention in a strictly controlled environment with inclusion and exclusion criteria, whereas operational research assess effectiveness within routine, and real-life settings.

Recently Zachariah and colleagues (2009) defined operational research as: “The search for knowledge on interventions, strategies, or tools that can improve the quality, effectiveness, or coverage of programmes in which the research is being done”.

Operational research involves descriptive, case–control, and cohort analysis. Some say that basic science research and randomised controlled trials is not operational research. However, effectiveness trials refer to whether an intervention works in people to whom it has been offered, and should in my view form an integral part of operational research. Results from such randomized trials can be are translated to benefit in the diverse setting of routine care.

For a health programme, the relevance of such research is whether it contributes to an improved performance or influences policy change at district, national, or even international levels.

Some examples of operational research from south Ethiopia include:

  • Antiretroviral treatment in resource limited settings (Jerene et al 2006): This cohort study assessed feasibility and effectiveness of antiretroviral therapy by use of historical controls.
  • An effectiveness trial in south Ethiopia aimed to evaluate if community health workers could improved smear-positive case detection and treatment success rates (Datiko and Lindtjørn, 2009). The study showed that involving of health extension workers (HEWs) in sputum collection and treatment improved smear-positive case detection and treatment success rate, possibly because of an improved service access. This finding has policy implications and could be applied in settings with low health service coverage and a shortage of health workers.

References:

Zachariah R, Harries AD, Ishikawa N, Rieder HL, Bissell K, Laserson K, Massaquoi M, Van Herp M, & Reid T (2009). Operational research in low-income countries: what, why, and how? The Lancet infectious diseases, 9 (11), 711-7 PMID: 19850229

Jerene D, Naess A, & Lindtjørn B (2006). Antiretroviral therapy at a district hospital in Ethiopia prevents death and tuberculosis in a cohort of HIV patients. AIDS research and therapy, 3 PMID: 16600050

Datiko, D., & Lindtjørn, B. (2010). Cost and Cost-Effectiveness of Treating Smear-Positive Tuberculosis by Health Extension Workers in Ethiopia: An Ancillary Cost-Effectiveness Analysis of Community Randomized Trial PLoS ONE, 5 (2) DOI: 10.1371/journal.pone.0009158

Integrating education, research and health care in developing countries

Posted on by — No Comments ↓

Models on how to integrate health service and research varies from country to country. Recently Dzau and colleagues from Duke University wrote about the experiences of using academic health science to transform medicine. They write that 5 billion people living in developing countries have inadequacies in hygiene and economic development, and health-care access are the main causes of shortened life expectancies.

They write that academic health science centres (previous medical schools) should play an important role promoting health and economic development. New organizational forms might improve health service delivery. By integrating health services, education and research and making this a collective responsibility it is possible to transform medicine, improve health, and reduce health-care disparities.

In most developing countries there is a strict division between universities and public health service providers such as hospitals and community health programmes. Often the Ministries of Education own the universities and Ministries of Health own the health institutions. Thus, universities become places where students get their degrees, and the quality of training often lacks the practical and real-life touch. The little research that is done is often weak and does not influence practice or policy making.

I believe we need new organisational frameworks integrating education, service and research to solve the huge challenges facing health in developing countries. Such an organization, that could include external partnerships, need to set research priorities, and develop models of education, care delivery and community health programmes, and has potential to enable health transformation.

Dzau, V., Ackerly, D., Sutton-Wallace, P., Merson, M., Williams, R., Krishnan, K., Taber, R., & Califf, R. (2010). The role of academic health science systems in the transformation of medicine The Lancet, 375 (9718), 949-953 DOI: 10.1016/S0140-6736(09)61082-5

2009 report on “Reducing Maternal Mortality Project”

Posted on by — No Comments ↓

Deaths from maternal causes represent the leading cause of deaths among women of reproductive age in Ethiopia. Thus, in line with the Millennium Development Goal for maternal health (MDG-5), this health project aims to reduce maternal mortality among the target population by two-thirds by 2015.

Experience from other countries show that two conditions are needed to reduce maternal deaths: Staff should be able to carry out comprehensive emergency obstetric care, and these services should be available to and used by pregnant women.

Vision and aims of project

In this public programme, we work with the Southern Nations, Nationalities and Peoples’ Regional State (SNNPRS) Health Bureau of Ethiopia (RHB) to improve maternal health and reduce maternal and neonatal deaths among the target population. The target population for this project are pregnant women in the following administrative areas of south-west Ethiopia:             Gamu Gofa Zone, South Omo Zone, Basketo Special Woreda, Dirashe Special Woreda and Konso Special Woreda.

The Project works with two levels of health institutions responsible for delivery services. Health extension workers are responsible for the kebele antenatal work, and hospitals and health centres are responsible for delivery services in the woredas and zones.

Our work has four components:

  1. Train non-clinician physicians (health officers) and midwives to carry out comprehensive emergency obstetric care (see
  2. Equip institutions to carry out comprehensive obstetric services
  3. Make delivery services available through health extension workers to all local communities and thus to pregnant women among a population of 2.6 million people.
  4. Using a simple, cost-effective, and sustainable tool to monitor maternal and newborn deaths. These community-based birth and death registries use health extension workers to register all births and deaths that occur in rural communities

Work in 2009

During 2009, 10 health officers, 10 anaesthetic nurses and 10 scrub nurses received training in Arba Minch Hospital. They now work at their home institutions. It is encouraging to see the these teams of health staff at Kemba and Konso Health centres, and Chencha, Saula, Gidole and Arba Minch hospitals routinely do emergency obstetrics, including caesarean sections. In November another four health officers and anaesthesia nurses started their training. In addition, we have trained about 150 HEWs and 30 midwives and clinical nurses.

Our project represents the first try In Ethiopia to train non-clinician physicians on a larger scale, and we are encouraged to see that comprehensive obstetric care is done at health centres in Konso and Kemba. In 2009, the number of caesarean sections increased by almost fifty per cent among our target populations, and the number of institutions routinely doing emergency obstetric care increased from two to seven.

Monitoring of work

As in many other African countries, Ethiopia lacks information on how many mothers die before, during or after delivery. Thus, by involving staff from regional health authorities, universities and health colleges, we have developed tools for community based birth registries. In 2009 we carried out pilot studies, and validated the tools to register births and deaths. In December we started birth and death registration for the population in Dirashe Special Woreda. This registration will enable the project to oversee if maternal deaths are reduced by two-thirds by 2015. Two master students now study at Gondar University, and one PhD student shall soon start at the University of Bergen.

We use experienced staff to follow and support the health officers at the rural institutions. In addition we continuously review the quality of the work at all institutions. So far, the results are encouraging and are comparable similar work started in other African countries.

Priorities for 2010

In 2010 we shall continue to strengthen the institutions, and through our Quality assurance, we systematic monitor and evaluate the work to ensure that standards of quality are being met. In 2010, our main emphasis shall be to strengthen the capacity of health extension workers, health posts and smaller health centres. The goal is to improve institutional birth coverage and that pregnant women in need of institutional care are referred in time.

More information is found at:

http://www.lindtjorn.no/page1/page11/page11.html

http://bernt.w.uib.no/my-research-areas/reproductive-healthproject/reducing-maternal-and-neonatal-mortality/

http://bernt.w.uib.no/training-programme/

Influenza confusion

Posted on by — No Comments ↓

For years, scientists have warned that an influenza pandemic, similar to the 1918 Spanish flu, might cause a global catastrophe.

We will remember the 2009 H1N1 pandemic more for causing confusion than catastrophe.

Although the new H1N1 virus is dangerous for the young and for pregnant women, in otherwise healthy people it causes a disease no more severe than seasonal flu. Up till now, the new H1N1 virus eventually showed more weaknesses than strengths in the world’s ability to combat such a pandemic:

  1. The World Health Organization held lengthy debates about using the name “pandemic”.
  2. When the virus was discovered, it had been circulating in humans for months. It was thus too late to control its spread with quarantines and antiviral drugs.
  3. Regrettably, many countries began useless travel bans and quarantines.
  4. WHO used consultants from the influenza vaccine industry. Only after recommending on large-scale vaccine production, did the public learn about their conflict of interests.
  5. The predictions about the extent and timing of the epidemics were poor. And, some governments, for example in Norway, scared its population by telling them that thousands would die.

Nevertheless, the biggest wake-up call has been to develop and produce vaccines on a large-scale. Unfortunately, it came too little and too late, and was only available during the pandemic’s second wave.

Each country should now review what happened. If this pandemic might serve as a trial run for the vicious killer that may come one day, there might be some important lessons that we learned.

Research on reducing maternal and neonatal mortality in south-west Ethiopia

Posted on by — No Comments ↓
Ethiopia is among the countries in the world with most maternal deaths. As part of our project to reduce maternal deaths, we have started several studies to get the necessary information to follow and improve our intervention. 

Monitoring maternal and neonatal deaths
We aim to set up a simple, cost-effective, and sustainable tool to monitor maternal and newborn deaths in a remote part of south-west Ethiopia. We shall set up a community-based birth and death registry using health extension workers.

Data from this research will help us to monitor the intervention programme to reduce maternal and neonatal deaths. The registry is a model for Ethiopia, and may also help other countries to set up birth registries.

We shall use, and compare several designs such as population based registries, direct demographic models (surveys) and institutional registries to measure maternal and neonatal mortality.
The research is collaboration between Ministry of Health in Ethiopia, Arba Minch Hospital, Gidole Hospital, Arba Minch University and University of Bergen.

Developing training programmes for health officers
Through the programme to reduce maternal and neonatal mortality we train non-clinician physicians (health officers) to carry out comprehensive emergency obstetric care. We regularly evaluate the outcomes of the operations they do.

On a separate web page we have outlined

Later, we shall also write about our experiences in setting up emergency obstetric services, at health centres and small rural hospitals. This will also outline the equipment needed to carry out such work.

A model rural hospital

Posted on by — No Comments ↓

There are unfortunately many hospitals in Ethiopia and in Africa that do not work as expected. They lack staff, or equipment. Often they lack staff doing essential interventions such as caesarean sections.

Gidole Hospital is a district hospital in south-west Ethiopia. It was a busy mission hospital. When the expatriate staff left, the hospital more or less collapsed. There were fewer patients, and patients had to be transported to a referral hospital for emergency surgery.

Now the hospital works as a rural hospital again. Two dedicated health officers do essential obstetrics, including caesarean sections. They also do essential surgery such as management of fractures, and treatment of severe wounds. These improvements have also brought about other changes: more patients with other diseases use the hospital and patient revenues has increased.

In my definition, a rural hospital should:

  • have a good antenatal programme and be able to routinely do emergency obstetrics, including caesarean sections
  • be able to handle wounds and common fractures
  • be able to diagnose and treat common infections in paediatrics and internal medicine
  • should have tuberculosis and ART programmes
  • have a good relationship with the population in its catchment area

The hygienic standard at the hospital is acceptable. The floors are often washed, and the patients receive clean bedclothes. In addition, the hospital should be able to make enough income to sustain work and keep staff.

Gidole Hospital does not have a doctor now. It works adequately with non-clinician physicians. 

Some lessons learned

Posted on by — No Comments ↓

To be able to reduce maternal mortality, two conditions should be met: Staff should be able to carry out comprehensive emergency obstetric care, and these services should be available to and used by the pregnant women.

About six months ago, we started to do caesarean sections at Saula Hospital in the inner part of Gamu Gofa. About 800.000 people live in these remote mountains.

Saula Hospital is a new hospital, which had not done any surgery before. We trained two operators (four months), two anaesthetist nurses (three months) and two scrub nurses. In addition we equipped the hospital with surgical instruments, an oxygen concentrator, suction machines and resuscitation equipment. Two experiences staff from Arba Minch Hospital taught the hospital staff how to handle and sterilize surgical equipments.

Our experience shows:

  1. It is possible to start emergency obstetric services, including operations such as caesareans sections and repair of uterine ruptures at places such as Saula. Non-specialists did the operations.
  2. The midwives correctly use partographs.
  3. Our review shows the indications to do surgery were correct. I believe the operations have saved many lives of mothers and neonates.
  4. Many women have severe complications already at admission to the hospital. This explains the high CS rate of about 20 %. It underlines that pregnant women in this remote part of Ethiopia come late for treatment.
  5. The number of uterine ruptures is high.
  6. Because of the late treatment, several of the women have developed vesico-vaginal fistulas.
  7. The complication rate for this newly started hospital is higher compared with operations done by non-clinician physicians at well-established hospitals. This underlines the importance in developing good and sound routines to ensure safe surgery.
  8. In our programme we review all operations, and we use a no-blame strategy to discuss complications.

One of the important lessons from Saula Hospital is to extend training the operators and anaesthetist nurses to five months at places where they shall start to do emergency operations. We also believe it is important to support and supervise such institutions for some years.

Now that Saula Hospital has set up the surgical and delivery services, emphasis must be to train midwives and nurses from the remote health centres to refer women to the hospital at an early.

Treating children with AIDS

Posted on by — No Comments ↓

We are reviewing the treatment results among patients receiving antiretroviral therapy at Arba Minch Hospital.


From Ethiopia there are few results published on treatment outcome in children with HIV infection. We have therefore reviewed the survival of children 14 years or younger.

The preliminary results (se Figure) show the five-year survival rate is about 90 %.

Most of the deaths occur during the first twelve months, and usually occurs among patients with advanced disease when we started treatment.

Long-term outcome of HIV treatment

Posted on by — No Comments ↓

We are reviewing the outcome of antiretroviral treatment at Arba Minch Hospital. We started treating patients in 2003, and since then about 1550 patients have received the antiviral drugs.

Our early papers showed the death rates among patients were high during the first months of treatment. We thought the reasons for these high death rates were the patients came late for treatment; that is, they presented themselves with late stage disease.

Our objective for the study is to see if the treatment results improve over time, and if patients present with earlier stage disease. Dr Zewdie Mulissa and Dr Degu Dare take part in this study.

Our preliminary findings confirm improved survival, and that patients come earlier for treatment. We are now also trying to find out what happens to patients after they are diagnosed with HIV until they start treatment.