Main wiki page | Recent additions | Recent changes | What links here | Categories | Category cloud
How-to guides | Organisation profiles | Project profiles
 

edit this page

CleanAirSIG:How useful are health questions in IAP questionnaires?


Table of Contents

CleanAirSIG
Established in April 2007, CleanAirSIG connects all those engaged in Clean Indoor Air and householdenergy in developing countries. CleanAirSIG is made possible with the kind support of the Shell Foundation
Discussion Topics
CleanAirSIG is sponsored by
Shell Foundation
SIG Sponsor LogoShell Foundation image

[top] [end]Introduction

Practical Action has just completed a study looking at ways to enable substantial numbers of people to access smoke alleviation interventions in order to improve their health and those of their children through addressing the barriers that prevent people accessing such interventions. This short document puts together some thoughts about health issues that seem to be overlooked because they are not the major ones.

[top] [end]Project Details

The study looked at the impacts on those who installed interventions to find out whether they were effective, and the social impacts of those changes. Because of the nature of the study, a ‘control group’ was not deemed possible.

In Nepal, virtually all those houses opting for interventions opted for smoke hoods, wall insulation, and small, but useful, changes to the cooking tripod to direct more heat to the pot and aid combustion. The reductions in smoke averaged around 80% reduction across around 180 households that were monitored.

In these previously highly polluted households, reported improvements in health were very substantial – see Table 1. As it was not possible to report over a full year, only the 71 households that had installed interventions for at least 9 months and were interviewed both pre- and post-intervention are included in this discussion. The total cost and total number of visits to the health provider are given below.



The major health impacts are discussed in detail in the book ‘Smoke, Health and Household Energy Vol 2.’ – just published. This discussion is just about the responses to questions about day-to-day health: Do you usually have a cough / bring up phlegm / sound wheezy first thing in the morning and at other times?



It would be easy to dismiss these findings as the responses of people ‘wishing to please’ – but a number of factors mitigate against this being the whole story' such as:
  • People paid for this intervention through micro-credit – and the smoke hood etc. cost around the same as for a buffalo – a substantial sum of money. Households were very willing to complain about anything they did not like.
  • Prior to the work, the fire was lit in a room that was almost completely closed, with smoke escaping through a small window, and sometimes through the door. Those visiting found it difficult to remain in the room for long - following the project the room felt clear and welcoming – so we should expect major changes
  • We know that smoke causes irritation of the airways and that carbon monoxide causes headaches etc. – it is not surprising that people who have lived in this situation are fully aware of these changes
  • Headaches – there is very little reduction in the number of cooks reporting headaches – though the frequency shows a noticeable reduction in frequency as in Table 3. Just over half the cooks still reported headaches as fairly strong.
  • The project adopted a highly participative approach – so those involved knew that we were trying to learn from what they told us – they felt part of the information team, and recognised the need for accuracy.


[top] [end]Discussion

What this information is showing us is that for those who had a smoke hood for nine months or over, according to the cooks themselves, around 90% of all those monitored suffered symptoms of cough and wheeze prior to the interventions most days, and that around 80% had a phlegm problem. This came down to 35% for coughs following the intervention, and to around 15% for phlegm and wheeze for the same criteria.

Before the intervention, most people had headaches, more than half of those suffered around half their lives with them. Following the intervention, this went down to less than once per week for all but two people – though people still said that they got headaches. How can we be sure that these are real changes, and not those brought about by household members being affected by the research itself (see, for example http://en.wikipedia.org/wiki/Experimenter_effect ), or by confounding factors?

If we use the conventional wisdom of monitoring the smoke, and predicting the changes in health brought about by smoke alleviation, it is only possible to get an idea of the major illnesses such as ALRI or COPD.

The health lobby is necessarily cautious in ascribing benefits to actions unless all the confounding variables have been considered, and unless there is good scientific proof (usually involving medically-trained staff) that what is happening is a direct result of the actions taken etc. Large sample sizes are needed and major and expensive research has been done for ALRI. But should the responses of individual people saying how they feel on a day-to-day basis not be equally valid? Obviously, there could be many other reasons for these changes – but when we know that smoke causes a lot of problems, then these findings should carry some weight.

There are many arguments that can be made against the study report such as:
  • Sample size is low,
  • the method is ‘unscientific’ and
  • the results are certainly not ‘evidence’.
However, everything here is based on how cooks themselves say they feel. It does not include long-term and serious illness – all it says is people say they are often ill prior to interventions, and much less ill afterwards.

So the question is 'Should we give give credence to what they say‘?

If the answer is no then without huge amounts of funds, how can we properly capture the real magnitude of ill health, and understand how it is preventing those living in poverty from improving their quality of life?

[top] [end]Some Key Points

  • If it is unacceptable to use the data from such surveys, should we be asking it? – or are we wasting peoples’ time?
  • Are there better ways of analysing our data to reach not only the major illnesses, but also to reflect the real ‘quality of life’ changes brought about by not having clean air? Could anyone factors based on hard scientific data to give values in days for the loss in time caused by non-life-threatening illnesses?
  • Is there enough qualitative evidence out there to allow us to say, at least tentatively, something about how day-to-day health is being affected by smoke alleviation?
  • When we are constructing a programme, do we need to factor in the ‘style’ of the project in order to give greater credence to the findings? (for example: those who pay for interventions are more likely to say there are problems than those for whom they were donated) – and if so, how should this be done?

[top] [end]Summary of e-discussion

Several issues were raised, but the majority of responses resulted from the question:

If it is unacceptable to use the data from such surveys, should we be asking it? – or are we wasting peoples’ time?



[top] [end]Key outputs from discussion

This discussion took a completely different route to that which I had expected. The important outcomes I felt were as follows:
  • Questionnaires should be treated with caution; for some people they could be an invasion of privacy, and therefore health problems could be under-recorded – for others the project itself could be seen as a route towards some sort of gain, and problems thus over-reported. However, they are useful to get a good understanding of general health issues in the area, the confounding factors and some clarifications about why certain things are hidden. If used, every opportunity to engage people in the process should be taken if they are to respond accurately – both the households and the team. Both Jay and Karabi felt that there were too many pitfalls for them to be viable.
  • Karabi and Neeta described a way to look at exposure based on counting the numbers of macrophages in a sample of sputum. Liz mentioned a non-invasive technique for measuring blood-CO.
  • There is a major study in Latin America which does not seem to be as well-known as it deserves. This may be because only the key findings are in English – though the full report is available in Spanish. A full report on a Healthy Kitchens project will be available early 2008.
  • Coming from this discussion – the work of Dr Accinelli was mentiond. His work has indicated that being exposed the first five years to smoke makes it 10 times more likely to develop COPD as an adult.
  • A long complex discussion on COPD followed – agreement was reached that the longer term effects of reduction in biomass smoke inhalation need to be further studied, as some of the early benefits appear to be very short lived. Tobacco smoke may not be a good proxy for biomass smoke in this instance.

[top] [end]Discussion in detail

Dr Jay Smith M.D., M.P.H., has just finished a smoke project looking at 44 households in the Inkawasi district of Peru. He did not feel that the health information was useful as results showed severe respiratory symptoms at the start of the project, but very few at the end – even in those households where the interventions had not proved successful. He felt that it was due to people wanting to believe that what they had done was successful – both project staff and household members. He commented that the only health results recorded in the final draft report were lung function (spirometry) results. Jay’s results (from his draft final publication) showed that when the households were split into two groups by the level of particulate reduction, the exhalation force showed an improvement in those households experiencing lower levels of smoke following the intervention. Removing a subject whose cooking status .changed and an outlier, made the difference significant (p less than 0.01). Tests that didn’t quite meet clinical validity criteria showed the same trend.

Magi Matinga flagged up the opposite side of the problem – people who are used to coughing all the time will say that they are not in poor health because it is such an everyday thing. This is even more complex where a language does not have a word for ‘clean’ – it is a concept which does not exist. They may also be unwilling to admit to a long-term cough in districts were HIV/AIDS is stigmatised – particularly if a local translator is involved – people may want to keep their health private. Despite these problems, Magi felt that health questions are valuable – particularly to get a good understanding of general health issues in the area, the confounding factors and some clarifications about why certain things are hidden. She feels that on the whole, one needs to use every available tool, gadget and method to pin point the answer, as people may be reluctant to speak of their problems, on the one hand, or try to please the interviewer if they feel that they will derive some monetary or physical benefit on the other.

Klas Heising responded to Jay Smith’s letter by flagging up the health impact investigation done by PAHO and the Cayetano Heredia University in Ayamachay, Inkawasi in 2003, headed up by Dr Accinelli. He wondered why these results had never been more widely publicised. Klas indicated that the same stoves, monitored by Aprovecho Institute, in both Peru and Bolivia, had proved very effective in reducing IAP and fuel. Liz Bates asked if the full paper was available in English as an abridged version (reported in Boiling Point 51), based on the key findings, was very interesting. Klas replied that the key findings are available on the HEDON network in English, whilst the full report is available in Spanish. pdf file link IAPmeasuringPERUgtz2005.pdf (3,476 KB)

Rogerio de Miranda responded to say that the work done by GTZ-PAHO in Inkawais has never been forgotten. It will be recognised again when the full report of the Healthy Kitchens project comes out in early 2008. Because of the success of the stoves, USAID-Winrock decided to scale it up and improve it. They are really happy that the methods and technologies developed for Inkawasi are spreading throughout Perú and elsewhere.

Karabi Dutta explained that during the work done by ARTI, their experiences with the health questions were very similar to those described by Dr Jay Smith, but that they did not do lung function tests as spirometry proved to be too problematic. When it was made clear that the interventions were not being provided for free, the initial headaches, back aches, watery eyes and dizziness disappeared as they did not want to buy the improved stove. In those households using the stove, many reported no change in condition as an excuse to avoid paying the monthly instalments.

In a new project Karabi intends to use a non-invasive respiratory health test, along with spirometry conducted by a medical team. In this test airway macrophages are obtained from healthy children through sputum induction, and the area of airway macrophages occupied by carbon is measured. This test, along with spirometry and room particulate monitoring is reported in http://lib.bioinfo.pl/pmid:16822993 . Liz asked about the complexity of using macrophages; do they need skilled medical staff and advanced laboratory techniques? – and also highlighted another non-invasive technique which provides readings on blood-CO through a clip applied to the finger http://www.masimo.com/rad-57/ . Neeta Kulkarni, who had introduced the concept to Karabi, stated some skills are required and a person from medical background is essential to supervise. This enhances the quality of the data, particularly for the lung function tests. The project can be easily done in India by involving a medical school and post graduate students.

Liz Bates feels that more weight could be given to the health questionnaire findings in their project because of way the project had been run, people felt engaged in the whole process – including the households. Their health status was very evidently improved (children without ‘runny noses’, with cleaner faces, less coughing and spitting, clothes not smelling of smoke) and she felt that the methods to look at just the serious ailments could risk underestimating the improved quality of life of those households who were part of the research – hence the question 'Can we adopt approaches that engage people so much that we are very confident that what we are told is accurate?'

Dr Niels Chavannes, of the International Primary Care Respiratory Group, joined the discussion to say that one cannot expect to see any improvement of lung function in COPD by reducing the levels of smoke. Any of the lung tissue that has been damaged by COPD does not regenerate - it is lost for good. The very best one can expect is to stop the situation getting worse. By reducing smoke in the kitchen, one is hoping to reduce the shortness of breath and coughing experienced by women with COPD, and thus to improve their quality of life. This does not mean that the COPD has improved, just that the condition has been stabilized and their environment makes it easier to breathe.

Klas Heising agreed - since COPD is irreversible in persons above 30years old. He highlighted the work done by Dr. Accinelli, which has focused on persons below that age comparing symptoms with and without improved stoves. Dr Accinelli states that being exposed the first five years to smoke makes it 10 times more likely to develop COPD as an adult.

Jay Smith responded that the tissue damage that Niels Chavannes was referring to is what is known as emphysema, which makes the lungs less elastic and thus less able to blow. However, COPD also causes inflammation, and thus narrowing, of the airway. When it gets bad enough, it is called chronic bronchitis – but before long-term damage is done, reducing smoke can lead to reduction in inflammatory narrowing and consequent improvement in airflow. Because the inflammation and the tissue damage are both caused by smoke, measuring the improvement in the first month after reduction in exposure can reassure us that the tissue damage should also be less on an ongoing basis.

Niels Chavannes responded that he felt that the variability in FEV1 measurement between people tends to be greater than the improvements measured in the short term.Data from lung health studies have indicated to the international COPD research community that a 3-5 year time frame is needed to find out anything about those whose lungs rapidly decline.Niel’s concern is that the diagnosis in some indoor biomass study subjects actually is not COPD but hyper-reactivity or asthma, which is a reversible disease. This is in fact good news for the subject (because their lung function may be greatly improved using available therapies), but may affect the validity of study results. COPD is by definition not reversible, as the internationally widely accepted guidelines have indicated over the past few years (see www.ersnet.org/COPD and www.goldcopd.com ). A permanent loss in lung function demonstrated by repeated and valid lung function measurement in a stable situation confirms COPD diagnosis, and any treatment should be aimed at improving exercise tolerance and quality of life by reducing shortness of breath, cough and exacerbations. Improving lung function has been discarded as a primary outcome measure after all the large therapeutic studies (Euroscop, Copenhagen City, BRONCUS, Isolde, Tristan) showed NO effects on longterm lung function decline. A short initial improvement during the first 3 months, which did not influence the slope of decline and thus was not clinically meaningful, was however noted in most studies. Even after smoking cessation, the ongoing inflammation process has been demonstrated in studies by the group of Emiel Wouters, showing the irreversible damage done, which can at most be modulated but sadly not reversed. Niels provided links to a paper which looks at this in more detail (Local and systemic inflammation in patients with chronic obstructive pulmonary disease: soluble tumor necrosis factor receptors are increased in sputum - Am J Respir Crit Care Med. 2002 Nov 1;166(9):1218-24).

David Whitfield flagged up an article in the New York Times that might be more accessible to those of us without medical training: http://www.nytimes.com/2007/11/29/us/29lung.html?i=5087&em=&en=f3ab09f50495cb72&ex=1196658000&pagewanted=all. It defines some of the jargon – so very useful.

Jay responded by explaining that the differences between Niels and himself are small, but that there is a well-documented improvement in FEV1% after smoking cessation. This is due to reversal of inflammatory narrowing and is a one-time effect. The more important long-term effect is the slowing of the loss of elastic recoil; this is also well-documented. He surmised that if this is the same as it is in smokers, maintaining "non-smoking" status with the same decrease in IAP over a longer period should also lead to less illness from COPD in the longer run. However, the reduction in inflammation can provide a good marker that the level of pollution reaching the lungs has been reduced. Jay pointed out that most of the households in which the work was done were not suffering from full-blown COPD.

Niels agreed that to study slowing of lung function decline effects, longer studies would be required, but put in a plea that if people have only symptoms which will lead to COPD, then it should not be called COPD. Rather that the symptoms and quality of life measurements should be monitored….but not called COPD.

Professor Jon Ayres expressed the view that the observations Jay records on change in lung function following removal of tobacco smoke exposure are small but measurable, but are also likely to be short-lived (California, Dublin and Scotland - the Dundee paper). In Dundee, lung function data was difficult to assess partly because of poor quality at the one year follow up, (the same problem occurred in Ireland), but symptoms improved at two months only to worsen slightly by one year (with no change in personal smoking status). As a result, Jon feels that Jay’s information on lung function is correct over the short term, but care is needed in extrapolating to a longer term. What is needed is a longer term follow up to be sure about how much inflammatory airway narrowing has been improved. A recently published study shows a clear relationship in the findings from a well-respected respiratory questionnaire (SGRQ) and tobacco smoke exposure in patients with COPD.

Liz Bates asked for some clarification for the lay-person. Is this all to do with the definition of COPD? Jay replied that for people who have smoked and have COPD, there is evidence that airway narrowing may even worsen for some despite smoking cessation. However, there is ample evidence that people who give up smoking loses less expiratory airflow (FEV1%) over the following years as compared to continuing smokers and that this may be due to slowing down the of loss of alveoli (air sacs). Jay believes that IAP causes inflammatory narrowing and destroys alveoli (air-sacs)through the same inflammatory response. Different species of particles may be involved with the two processes and that the proportions of these IAP particles may not be the same as those in tobacco smoke. If this is the case, it would be necessary to test for COPD changes with biomass smoke as tobacco smoke may not be a good proxy. Jay is concerned that if we take what is known from the industrialised world COPD surveys, the results for the developing world could be biased by our expectations. Symptom related quality of life is hugely important but some of the worst lung function results in their study did not seem to be bothering people much. It may be because there were no smokers and only a handful of people with mild COPD. Jay feels that mild COPD is often undiagnosed, and the loss of function , is so gradual that people aren't complaining about it asthma and greater improvement in those with asthma.

A final comment was that he felt that it was important to reward people for taking part. This opens up a whole new debate! He also felt that the studies should be done in private, preferably at a health clinic where people could also be made aware of the links between health and environment.

[top] [end]Contributors

User: Liz Bates November 2007

edit this page

Page created: 16 November 2007; Last edited: 28 July 2008; Version: 9
Wiki text is available under the terms of the GNU Free Documentation License.

Pagename: CleanAirSIG:TheUsefulnessOfHealthQuestionsInIAPQuestionnaires @HEDON: KPHA