Tropical Health Matters

Philippa West and colleagues have demonstrated yet again that ownership of an insecticide treated bednet (ITN) does not guarantee that people are protected. Their study in Muleba District, one rural district in northwestern Tanzania points out the universal challenges of universal coverage.

In the study district the proportion of households (HH) owning ITNs increased from 63% to 91%.  The average number of ITNs per HH also rose from 1.2 to 2.1. The problem was that even with more nets in more homes, the proportion of residents actually sleeping under them did not rise to reach the desired target of at least 80%.  As an aside – statistical significance was achieved, but not program significance.

Here is what they found in the community from before and after universal coverage distribution:

• proportion of all residents sleeping under ITNs rose from 41% to 56%
• children under five years old – 56% to 63%
• pregnant women  – 55% to 63%

What was happening? One crucial finding was that 42% of households and fewer nets than the number of sleeping spaces and 20% had more than enough.  This speaks poorly to pre-distribution planning and HH registration or lack of care and verification during the actual distribution.

Another explanation documented by the researchers was that a fair number of HH did not redeem their net coupons at the distribution point – too busy, forgot, no transportation, etc. The program could have benefited from community directed distribution which guarantees that villagers take care of and ensure their own supplies of basic health commodities like ITNs.

These findings do not come as a surprise. Numerous reports from Demographic and Health or Malaria Indicator Surveys have shown a similar phenomenon – ownership of a net by the household does not guarantee that people actually use them.  In particular we can see this problem in the attached chart from the Tanzania 2010 DHS.

Fortunately the distribution in Muleba, though having problems, was equitable in terms of the economic status of recipients. Better planning, health education, community involvement and follow-up is required  if we are not to waste millions of dollars, not to mention lives, from poorly distributed nets.

Efforts to eradicate smallpox and guinea worm have taken generations.  In both cases there was a very clear and focal transmission pattern. Smallpox spread only among people and could be stopped with a very effective vaccine. Guinea worm again only infects humans and transmission can be stopped through safe water.

Unlike these other diseases malaria has no one silver bullet and transmission dynamics vary across many different environment types.  At present case containment that was successful in ending smallpox and is effective in guinea worm, is out of the question for malaria.  Malaria must deal with huge health systems challenges ranging from weak procurement and supply management systems to health workforce shortages.  Peak efforts at malaria control have also unfortunately coincided with a world economic downturn.

Documentation of malaria control progress is ongoing, if not perfect. A look at indicators from three national DHS/MIS surveys in Uganda make it possible to show how difficult it is to achieve and sustain coverage of the interventions we do have. To date the Roll Back malaria targets of 80% have not been achieved for any indicator, and in the cases of using insecticide treated nets (ITNs) and intermittent preventive treatment in pregnancy (IPTp), there have been drops.

There are a number of ways to measure indicators.  For example, the figures for people who slept under any kind of net are better than those using only ITNs. On the other hand, if we used the data on taking Artemisinin-based combination therapy (ACT) within 24 hours of fever onset, then the figures would be worse.  Of course these figures do not even include whether treatment occurred after a positive rapid diagnostic test.

What we can see is that even with a little more positive nudge, the data are not encouraging.  The guinea worm eradication effort has shown that stakeholders do tire of maintaining disease control efforts year after year.  Many endemic countries are still much too dependent on external assistance to go it alone in eliminating malaria. What will it take to get malaria control and elimination back on track so we can achieve zero malaria deaths by 2015?

Uganda’s 2010 Roll Back Malaria Roadmap seemed reassuring.  Apparently 2.7m nets were already in place by late 2009, and supposedly a supply of another 18m long lasting nets (or at least the funding) was ready for achieving universal coverage by December 2010.

Assessment of Uganda’s Roadmap progress credited the country with achieving procurement of these nets. It seems odd therefore that the 2011 Roadmap indicates that 6.4m nets are in place and 10.4m need to be distributed in 2011. What’s going on?

A new study by Carla Proietti and colleagues provides some answers. Not only do they document continued high transmission in the northern part of Uganda (polymerase chain reaction rate of 72% in children below five year of age), a situation that threatens control efforts by neighbors, but they also identify plausible reasons for the lag.

The researchers politely suggest that, “The failure to reduce the burden of malaria could reflect sub-optimal implementation of malaria control measures.” They also explained that, “Malaria control efforts in Apac (sub-county) were not reliably monitored in the last decade and affected by political unrest in preceding years.”

Stockouts of anti-malarial ACTs was also listed as a problem. It should be recalled a few years ago that, “The Global Fund has decided to suspend its five grants to Uganda because there is evidence of serious mismanagement by the Project Management Unit (PMU) for Global Fund grants in Uganda.” Although the programs have resumed, satisfactory settlement of the problem was not achieved.

In light of this study Childsurvival.net warns us not to let successes in recent years blind us to reality. “Those who believe that Africa is within shooting distance of malaria elimination may wish to reconsider their position after reading this article (Proietti et al.). One should qualify this Ugandan article in several ways: 1) Local insurgency in the area under consideration, 2) Hiccoughs with the GF over misappropriation of resources, 3) Autocratic gerontocracy at the national level. Unfortunately, these three factors are not peculiar to Uganda.”

As mentioned yesterday, Roadmaps are a good tool to help us plan for malaria control and elimination – but we must stay on the road for them to work.

The Roll Back Malaria Partnership guided countries to develop 2010 roadmaps for major malaria commodity and support service availability and gaps. The aim was to aid planning to reach universal coverage by the end of 2010.  Forty-seven countries/locations on the African Continent and surrounding islands completed the analysis and started moving down the road to success.

In the case of 36 countries the road became a little longer than anticipated.  Part of the challenge was international – there are only a few manufacturers of long lasting insecticide-treated nets, for example. Some of the barriers were internal, inadequate estimates of the logistical costs to distribute commodities, even if they were in hand. Now we have 2011 roadmaps in an effort to meet up with the original 2010 goals of 80% coverage with essential malaria commodities.

At least one-quarter of countries that actually targeted a specific intervention in 2010, did not meet the 80% goals.  Of particular concern is the fact that Rapid Diagnostic test use is both off target and not keeping up with ACTs.

Meeting procurement and distribution targets is one step, but getting people to use malaria control interventions is another challenge. As the director of a prominent Nigerian NGO recently said, “… ‘though about 35.6 million nets have been distributed across the country, it is highly under utilized,’ which according to him is responsible for the high death rate associated with malaria.”

Nigeria provides an instructive case. The roadmap for 2010 called for 62.9m LLINs of which 4.4m were already in place and pledges were set for 49.4m. This left a gap of 9.2m.  While the RBM 2010 roadmap analysis shows that Nigeria met its LLIN target, the implication is that the target did not include the gap.  Now the 2011 roadmap for Nigeria now shows that resources are in hand for both the 9.2m gap from the 2010 campaign plus an additional 8.2 m for routine distribution in clinics as a keep-up measure.

The gross figures do not fully reflect the fact that of the 36 states (plus one capital territory), campaign distribution of LLINs continued from 2010 into 2011 in 17 states. So far 9 or the 17 have completed distribution, but by carrying the campaign into 2011 additional delays were met in the remainder due to national elections, delayed local funding for the effort, and distribution logistics. So again while the roadmaps help identify commodity gaps, they do not always identify the challenges at the level of distribution and use.

The roadmap process is an important planning tool. It needs to be supplemented with plans for logistical support and health education to encourage use of the malaria commodities and services that are eventually distributed.  For example, Nigeria estimates that it needs close to $17m for Monitoring and Evaluation and Information. Education and Communication. We can see from the Nigerian roadmaps that this planning needs to be a continuous process – not only is annual resupply of ACTs, RDTs and SP for IPTp needed, but also continuous stocks of nets for routine, keep-up services.

We are unlikely to eliminate mosquitoes, according to Tanya Russell and colleagues, but she notes that this should not stop us from implementing all available interventions. Specifically their study of malaria vectors in Tanzania found that the at reduced densities of mosquito populations, they try to reproduce more, meaning we may never get below 10% mosquito elimination.

Instead, a member of the National Malaria Control Program in Tanzania says our goal “should be to reduce, and eventually halt, transmission of the parasite, rather than eliminating the vector.” If we can achieve no more than 90% elimination of mosquitoes, what is a realistic coverage figure for malaria interventions?

Applications of net and case management strategies in Rwanda and Ethiopia have definitely shown that major drops in malaria incidence are possible.  But the RBM targets of 80% coverage (85% for the US President’s Malaria Initiative) are elusive.  Demographic and malaria surveys from Senegal, Liberia and Nigeria show that even in homes that own nets, net use among people at most risk, does not reach this target.

Are we really sure that 80% is the right target?

Fred Binka was one of the first to demonstrate that people living in homes without nets can be protected by their neighbors’ nets, which kill mosquitoes in the community. ITNs “provided very good personal protection to children using them, and also protected nonusers in nearby compounds. Among nonusers, the mortality risk increased by 6.7% with each additional shift of 100 m away from the nearest compound” with nets. This led the researchers to speculate on the need to study whether the “mass effect from a small number of highly dispersed nets would provide equivalent protection to complete coverage.”

A few years later William Hawley and co-researchers reported that, “protective effect of ITNs on compounds lacking ITNs located within 300 meters of compounds with ITNs for child mortality, moderate anemia, high-density parasitemia, and hemoglobin levels.”

As part of the move toward universal coverage, Killeen and colleagues examined the importance of considering all household members, not just the ‘vulnerable.’ The group condluded that …

Using field-parameterized malaria transmission models, we show that high (80% use) but exclusively targeted coverage of young children and pregnant women (20% of the population) will deliver limited protection and equity for these vulnerable groups. In contrast, relatively modest coverage (35%–65% use, with this threshold depending on ecological scenario and net quality) of all adults and children, rather than just vulnerable groups, can achieve equitable community-wide benefits equivalent to or greater than personal protection.

Barat has called for ‘data driven decision making‘ in the effort to eliminate malaria. Using data in models as done by Killeen is a further important step. The onchocerciasis control community has been working with such models for over 15 years now. New data are fed into the Onchosim model based on program progress such that it is possible to forecast that onchocerciasis could be eliminated from areas with high initial prevalence if 65% coverage of ivermectin treatment were maintained for at least 25 years.

Unlike onchocerciasis control, malaria elimination rests on multiple interventions.  This makes modeling much more urgent, as outlined by malERA’s research agenda for eradication. Since universal coverage unfortunately does not mean universal usage, we need to seek valid data and models to help us plan for distribution of malaria interventions more strategically in ways that are affordable and can be maintained and at the same time can achieve maximum reductions in morbidity and mortality.

Use of long lasting insecticide-treated nets has an impact on malaria transmission, but the key to achieving an effect is ensuring that people both own and use the net.  A study from southern Benin has taken this process to the next level. Georgia Damien and colleagues found that, “only correct use of LLINs conferred 26% individual protection against only infection.”

The authors distinguished the use of LLINs – whether children were sleeping under it during the control – from the correct use – whether the LLINs were correctly hung and tucked and were not torn.
As reported by other sources, possession of a LLIN, in this case over 90% of households in a southern Benin community, did not guarantee use, which varied on average from 73% in the rainy to 67% in the dry seasons.  Correct use likewise varied from 68% to 42% by season.

Although the Benin study implies that the protective effect of nets may result only from ‘correct use’, earlier work in Ghana showed that some protection was possible even if a household did not have LLINs.

Binka and co-researchers reported that, “The death rate among unprotected individuals increased with distance from the nearest compounds with bed nets. This suggests that (insecticide-treated nets) are protecting other individuals without bed nets who sleep close to protected compounds.”

Much of our hope for achieving morbidity and mortality goals rests on what could be called this community protective effect of nets, since it is difficult, considering the nature of human behavior, to expect everyone to use a net, let alone use it correctly.

We now know our goal for universal net coverage will not be met until well into 2011. The challenge as can be seen in the photos is not just distribution, but effective community health education to ensure that nets are valued and used correctly.

In the waning hours of 2010 several public health goals and targets come to mind –

• Access to clean water and sanitation by 1990
• Guinea worm eradication by 1995
• Health for all by 2000
• Polio eradication by 2005
• Universal Coverage of malaria interventions by 2010
• Millennium Development goals by 2015

As of October 2010 Nigeria had distributed long lasting insecticide-treated nets (LLINs) 14 of the 36 states and the Federal Capital Territory and about 40% of the targeted 60 million plus nets. A key challenge was, “The lack of operational funds to support campaigns in 22 states significantly resulting in undue delays in the delivery of LLINs.” More progress was made during the remaining months of the year.

Where are we on 31st December 2010? The Roll Back Malaria Partnership’s most recent report on progress toward targets (the country road maps) is seen in the chart to the right. Intervention coverage progress is based on the number of countries that are actually implementing nets, medicines and spraying.

As can be seen the best progress comes with treatment and preventive medicines (ACTs and IPTp respectively).  Only 64% of countries have distributed at nets to at least 80% of the targets. The biggest gap in in the area of rapid diagnostic testing.

Distribution of an intervention does not mean actual coverage has been achieved. A recently reported study from Nigeria shows the challenges once nets reach the household.

Oyeyemi and colleagues found that 95.2% of households has received a net after a campaign. Unfortunately progress went downhill from there: “87.3% of the LLINs received were present in the households during the survey and 52.1% of households hung their LLINs … (and) utilization rate of a LLIN among the sampled population was 59% the previous night before the survey.”

Recent Demographic and Health and Malaria Indicator Surveys from places like Liberia, Senegal and Nigeria show that possession of a net by a household is not a guarantee that it will be used.

From the chart above we can see that efforts to attain universal coverage – or more accurately universal distribution – will have to proceed into 2011.  Distribution goals require health systems strengthening. The coverage goals will require more intensive community outreach and education to ensure these interventions are actually used.

We are achieving outputs – commodities distributed; we are struggling with outcomes – commodities used. What will we see in terms of impact by 2015 – the latest on the list of public health targets where we started this posting?

Tarry Asoka, a medical doctor and health development consultant, provides us a perspective on why health insurance is needed to meet malaria treatment gaps. Tarry is  Publisher/Editor of Health Insurance Affairs and Malaria Bytes…

All across sub-Saharan Africa the poor utilization of modern health services usually reverses and begins to improve, reaching a tipping point as soon as there is confirmed indication that ‘treatment charges’ in health facilities have been removed. And politicians in the sub region are very quick to take good note of this phenomenon – often taking full advantage to develop populist ‘free health campaigns’ that are often not sustainable.

The lack of continuity is not usually the result of faulty design but due to poor execution as many of these are mostly ad hoc initiatives rather than enduring programmes. But the fact that these campaigns continue to be popular especially among poorer citizens despite their lack of permanence and irrespective of who is organising it – government, NGOs or private – should give health planners some worry that something is not quite right.

Curiously, malaria is still the most common condition recorded by health professionals during such health jamborees. A recent free medical check-up drive to promote a new community-based health care programme in a high density area in Port Harcourt, Rivers State, Nigeria – noted that close to 30% of those who were seen had classical symptoms and signs of malaria that have not been treated for at least 2 days.

So what could have happened to such persons especially children if this event did not take place at that particular point in time? Your guess is as good mine.

But one fact is clear – the payments that are needed to be made at the point of accessing health services prevent large majority of the population from seeking medical care. A recent survey in Kenya, for example, found that “61.5% of individuals who did not seek (malaria) treatment reported that cash shortage was the main barrier.” Others coped by borrowing, selling household possessions, or buying cheap drugs from shops.

Therefore, any mechanism that enables people to access care ‘free at the point of delivery’ will improve treatment for life-threatening conditions such as malaria and save lives.

This is ‘no-brainier’, and does not require elaborate plans to be put in place. Apart from informal and community-based health insurance, which has been quite challenging to set up, other approaches such as vouchers and coupons have also proved to be useful alternatives.

The task now is to scale-up these options to achieve universal coverage.

There is hope among world leaders that the Millennium Development Goal of reducing malaria deaths to near zero is now likely to happen. Ray Chambers, the UN Special Envoy for Malaria is quoted as saying …

“Today, enough nets are in place to protect 75 percent of those at risk, and we will reach universal coverage by December 31, 2010, an astounding testimony to the power and efficacy of the unified global campaign. These nets have reached nearly 500 million people in the last two years alone, and their impact on saving lives is profound – current levels of intervention are saving 200,000 lives per year. We are on track to meet the Secretary-General’s goal of ending malaria deaths by 2015, and our work won’t be finished until we do.”

Having nets ‘in place’ and having nets used are two different indicators of success. ‘Nets in place’ will not achieve the MDG for malaria unless nets are used. Numerous surveys have been mentioned in our previous postings showing that even when nets are ‘in place’ in homes, they are not always used by the most vulnerable members of the household.

Another concern is that even if nets are ‘in place’ by the end of 2010, we may not achieve the MDG of near zero malaria deaths. LLINs are known to wear out after 2-3 years of normal use. Unless there are plans for net massive replacement efforts around 2013, we may see slippage in attaining and sustaining the goal.

A major weakness of past public health programming has been providing people with technologically sound and useful interventions without taking into full account the social, cultural and behavioral factors that influence acceptance and use of the interventions.  In order to continue to save lives with nets and other malaria interventions, we must strengthen the social and behavioral components of programming now and not wait until 2015 to see if we actually saved lives or not.

Guest Blog by Tony Anammah, DELIVER PROJECT, Nigeria

The Universal LLIN campaigns in Sokoto, Kebbi, Kaduna and Adamawa States had one thing in common – the LLINs were provided alongside other interventions like immunization (especially polio), nutrition screening, vitamin A supplementation, deworming and provision of ante natal care services.

A quick assessment of what transpired during the Child Health Week in these states showed that there was some level of success in integrating the interventions in the above metioned states. This success was in terms of the fact that communities were more willing to accept polio immunization because of the  ‘mosquito’ nets. The Immunization Plus Days (IPDs) actually provided a good structure for the delivery of these nets to households and the structure designed for the distribution of the nets (if it had been stand alone campaigns) was modified to incorporate all the interventions.

The challenge was not if the communities will accept the nets or ensuring that the nets gets to all the communities. The major challenge is if the households will actually use the nets. This needs to be closely monitored and followed up. There was an end process monitoring after the campaigns but it will be interesting to know how well the households are using their nets some months later.

Even though there were some successes, there were a number of challenges. One of such challenge was that the time to plan for such an elaborate campaign to integrate all the interventions was evidently too short and there were clashes in programmes delivery strategies. There were some level of cooperation between the programme managers but at the same time, each manager was equally keen on delivering on their individual programme obejectives.

On the average, it can be said that there was some level of success but  it will be wonderful to see if this kind of attempt to integrate interventions can be sustained and lessons learnt incorporated into strategies. But most importantly, it must be ascertained if integrating these interventions has actually been beneficial to the fight to reduce mortality due to malaria.