17 pages
Le téléchargement nécessite un accès à la bibliothèque YouScribe
Tout savoir sur nos offres


  • fiche de synthèse - matière potentielle : the stability of the different chelates
  • fiche de synthèse - matière potentielle : common chelate stability with different ions
  • exposé - matière potentielle : mineral element movement to root surface
Practical Hydroponics & Greenhouses . July/August . 2007. 39 MICRO-ELEMENTS in Agriculture EYAL RONEN outlines the importance of micro-elements, their availability problems and possible solutions through chelates. 23_PH&G_95.qxp 11/05/2007 3:15 PM Page 39
  • importance of micro-elements
  • stability constant
  • uptake
  • ethylenediaminetetraacetic acid c10h16o8n2 dtpa diethylenetriaminepentaacetic
  • metal ion
  • root
  • high value
  • elements
  • plant
  • soil



Publié par
Nombre de visites sur la page 28
Langue English
Signaler un problème

Andersson BMC Health Services Research 2011, 11(Suppl 2):S1
Building the community voice into planning:
25 years of methods development in social audit
Neil Andersson
Health planners and managers make decisions based on their appreciation of causality. Social audits question the
assumptions behind this and try to improve quality of available evidence. The method has its origin in the follow-
up of Bhopal survivors in the 1980s, where “cluster cohorts” tracked health events over time. In social audit, a
representative panel of sentinel sites are the framework to follow the impact of health programmes or reforms.
The epidemiological backbone of social audit tackles causality in a calculated way, balancing computational
aspects with appreciation of the limits of the science.
Social audits share findings with planners at policy level, health services providers, and users in the household,
where final decisions about use of public services rest. Sharing survey results with sample communities and service
workers generates a second order of results through structured discussions. Aggregation of these evidence-based
community-led solutions across a representative sample provides a rich substrate for decisions. This socialising of
evidence for participatory action (SEPA) involves a different skill set but quality control and rigour are still
Early social audits addressed settings without accepted sample frames, the fundamentals of reproducible
questionnaires, and the logistics of data turnaround. Feedback of results to stakeholders was at CIET insistence –
and at CIET expense. Later social audits included strong SEPA components. Recent and current social audits are
institutionalising high level research methods in planning, incorporating randomisation and experimental designs
in a rigorous approach to causality.
The 25 years have provided a number of lessons. Social audit reduces the arbitrariness of planning decisions, and
reduces the wastage of simply allocating resources the way they were in past years. But too much evidence easily
exceeds the uptake capacity of decision takers. Political will of governments often did not match those of donors
with interest conditioned by political cycles. Some reforms have a longer turnaround than the political cycle; short
turnaround interventions can develop momentum. Experience and specialisation made social audit seem more
simple than it is. The core of social audit, its mystique, is not easily taught or transferred. Yet teams in Mexico,
Nicaragua, Canada, southern Africa, and Pakistan all have more than a decade of experience in social audit, their in-
service training supported by a customised Masters programme.
Epidemiology as a living language between In the early 1980s, the Italian labour movement “alter-
people and public services nativa operaia” [77] put forward the idea of community
Over the last 25 years, several million members of the engagement in scientifically defensible epidemiology.
public and public servants in dozens of countries have Principles like validity of community views, collation of
community experience and validation through scientificparticipated in CIET social audits of health related issues.
Between 1994 and 2010, 45 health sector social audits in measurement contrasted the images of white coated
27 countries contacted 504,057 households [1-76]. Addi- scientists coldly observing “subjects” of research. In 1984,
tional file 1 summarises the topics, sample and main follow-up of Bhopal survivors developed sampling and
results. interview approaches that were robust and reliable in
developing country conditions. The practical linkages
Correspondence: andersson@ciet.org with national health agendas had their roots in UNICEF-
Centro de Investigación de Enfermedades Tropicales (CIET), Universidad
sponsored work in Nicaragua and Honduras in the mid
Autónoma de Guerrero, Calle Pino, El Roble, Acapulco, Mexico
© 2011 Andersson; licensee BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.Andersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 2 of 17
1980s, concerned with evidence on key child and mater- inferred from a battery of questions following some theory
nal health outcomes [78,79]. At that time, the incomple- base, and more about what people meant to express and
teness and inaccuracies of routine health record systems what the enquiry meant to them. At least as important as
in most developing countries all but eliminated the infor- the first order information about the indicator of health
mation value of the voluminous but patchy data. outcome, we found we generated second or third order
Concerned with the principles behind the Primary information of what engaged communities can understand
about the indicator and its determinants, what they thinkHealth Care ideal [80] but keen to avoid token community
can be done about it, and how that should happen. Just asparticipation [81], the Central American project viewed
we aggregated vaccination and costs of measles [36], skincommunity engagement as reaching beyond those who
used services. For much the same cost as the unreliable conditions [82,83] or seropositivity for Chagas disease
routine data collected, in this case on infant and maternal [35], we found we could aggregate community-led solu-
health outcomes across the country, we engaged a sample tions to those problems into a regional strategy.
of communities in a mix of qualitative and quantitative
research methods. We tried to optimise information con- Evidence and guesses in planning
tent and use of evidence through an inclusive approach Often defying the risks of reduction and over-interpreta-
that engaged communities and service workers. A cross- tion, day-to-day health planning is all about causality.
design of standard epidemiological and qualitative tools Worse, it is about projections – guesses – of causality.
measured common outcomes like diarrhoea and maternal Planning assumptions are often heroic: the vaccine will be
morbidity. We wanted to look upstream from these health kept and administered correctly; women will attend a pre-
outcomes to potential causes – health choices and use of natal clinic where they will receive what they need; doctors
health services. will get it right; medicines will be there and will work. But
The idea was not to blend qualitative and quantitative health services do not always work as expected. They do
approaches into some half-way method. We broke up the not reach all those who need them; they do not always
research process or, as in linguistics, we parsed it into dif- have the intended effect for those who use them.
ferent moments. Each of these moments had a distinct Health services are a live series of subsystems. Health
objective and method. A very participatory moment set workers have lives to lead, bills to pay, and all this influ-
the conceptual framework; a more technical moment ences health care where it meets the intended beneficiary.
fitted standard questions to this conceptual framework; in Social audit is a stocktaking of where we are with these
a tightly supervised cluster survey, interviewers read the assumptions, guesses and intentions. The idea is to pro-
questions and wrote the answers; an undemocratic data duce hard evidence about what works, who is left out
entry moment digitised responses to the questionnaire, andwhatwillmakeuptheshortfall.Whileafinancial
with no added value from the keyboard operator; analysis audit looks at how financial resources meet financial
(computation) was technical; a separate community objectives, a social audit looks at how resources meet
engagement component discussed the results and feasibil- defined social objectives. The core activity of stocktaking
ity of potential solutions, typically through focus groups in is to get evidence that tells us about health service perfor-
each cluster. mance. The original description in 1985 identified three
The output included multilevel (individual, household evidence types–“words in a common language” [78,79]:
and community) data that engaged stakeholders at each Impact is the change of status (number of diarrhoea
level. We worked on two simple principles. First, an epide- cases or a reduction in unofficial payments) attributed
miological sample of domains (usually communities) to a particular intervention;
could result in representativeness of the final evidence. Coverage is the proportion who receive a particular
Second, repeated cycles of measurement in the same sites service (such as bed-nets, vaccination or access to clean
could decrease random error of the measurement. water) out of all those who need it – not only out of
Although repeated visits bring other problems, the result those who access the services;
was a method to measure health service performance and Cost includes time, staff, cash, supplies, transport and
to understand and to use community engagement in all other elements required to supply or to take advantage
bringing about improvements. of a given service or programme. It includes the cost to
Behind our social audit approach is the idea of epide- service users as well as the cost of providing the service.
miology as an evolving and self-organising system, a lan- Linking these three types of evidence in their implicit
guage instead of a rigid tool, with increasingly informed relationship – coverage of the intervention, that causes the
community engagement increasing relevance of the emer- impact, at a given cost – gives meaning to public service
ging solutions. By engaging residents of the clusters or performance. This is what most planners want to discuss.
“sentinel sites” in dialogue about their answers to ques- A common failure of health information systems is that
tionnaires, the approach was less about the indicator evidence comes mostly if not exclusively from institutions.Andersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 3 of 17
For example, vaccination rates among children who attend study that achieved 93% five-year follow-up in the after-
a well baby clinic. Or maternal deaths among women who math of the infamous Union Carbide disaster in India
delivered in health facilities. Services that exclude some [84,85]. In the early days simply a school exercise book,
people by charging too much also exclude these potential this lined ledger has pages cut in half vertically. The
beneficiaries from institution-based information systems. interviewer reads the questions from the questionnaire
One or two percent missing randomly will not affect the pasted on the inside of the front and back covers and
big picturemuch. But some peoplefailtoturnupat health writes the answers on the corresponding line of a page of
facilities for reasons. If it is those who cannot afford health the book identified for each household, one household
per page. Separate pages can serve for different membersservices, or who have reduced access for reasons of culture
or distance, hospital or clinic data will be frankly mislead- of a household. The books are inexpensive, usually avail-
ing sources of evidence on the public health. Social audits able locally, robust and reliable in field conditions.
go to the population base. They find out what people need In the household survey, interviewers contact contigu-
and what they get, and relate this to the service offer. ous households in each cluster for statistical handling as
The simple fact that social audit goes to the population a mini universe. This reduces waste of time between
base opens another dimension. The simpler first order households but importantly allows for the interaction
product of social audit is evidence on use of health between households, for neighbourhood or place, as part
resources and on service performance. Deliberately enga- of the research process. We link these household data
ging the community, or even just “being there”, adds a with data from other sources in the same site: institu-
range of predisposing, enabling and engaging dimensions tional reviews of relevant facilities and qualitative data
that affect health and health service behaviour. This is from key informants and focus groups.
the real science of epidemiology in social audit: under- Preliminary analysis of this quantitative evidence pro-
standing, enabling and engaging dimensions and under- duces a first round of key findings for discussion in focus
stand how these might affect measurement, and how groups –inthesamecluster – to gain a qualitative per-
they might be part of the solution to whatever problem spective on the findings, particularly views on solutions to
the social audit measures. the problems. Thus, the household survey permits aggre-
gation ofdataon occurrence, such as diarrhoea, household
What happens in a social audit opinions of services, costs and so forth. Sharing these
Our 25 years of experience with this approach crystal- results with the clusters, we collect qualitative data on
lised a typical sequence of activities in two main phases, how to deal with these occurrences, in the same clusters.
summarised in Table 1. We almost always begin with a Institutional review of facilities serving the clusters might
include local analysis of routinely collected data, personneldetailed consultative process, to frame the issues, before
reviewing what existing data sources can produce on the issues, times of operation, costs of services and charges,
problem. The typical sample comes from the latest cen- and relations with the community. Some social audits
sus, although this is not an invariably reliable sampling have included a sample of health workers completing a
frame. standard questionnaire (like Procol). It might also include
A household survey usually follows, almost always with observing institutional water supply, curtains for privacy,
face-to-face interviews. The physical data collection or even the flow of patients and their treatment.
instrument commonly associated with our social audits is The leading epidemiologist analyses the layers of evi-
the “Bhopal book” (Figure 1). We developed it in an dence. The research team feeds these preliminary results
emergency setting to collect data from households in a into discussions of gender stratified focus groups, and
Table 1 The two phases of a social audit
Phase 1: design and data collection
clarify the strategic focus
analyse existing data to identify gaps and generate operational questions
design sample, instruments and conduct pilot test
collect information from households, institutions, and key informants in a panel of representative communities
link public service and data, analyse in a way that points to action
Phase 2: socialising evidence for participatory action
take findings back to the communities for their views about how to improve the situation
summarise information for policy and management (eg score cards)
evidence-based training of planners, service-providers and media
partnerships with civil societyAndersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 4 of 17
Figure 1 Bhopal book. A Bhopal book used during the Bhopal followup. The pages are cut vertically and the questions written on the cover.
with health service workers (Figure 2). Then these results taken about health-related actions, including use of pub-
make up the social audit product. lic services. Socialising research results involves two
Social audits target three types of research users: plan- feedback dynamics. First, within each sentinel site feed-
ners at policy level, planners in the health services, and back of findings generates a second order of information
planners in the household, where the decisions get – community led solutions. Second, fact-finding and theAndersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 5 of 17
Figure 2 Focus group. A focus group discussing access to health care.
action it leads to should hold influence beyond the was not to prove causality, but to take discussions one
.immediate site of data collection A statistically interpre- step beyond simple indicators and presumed causality.
table sample allows for aggregation of community-led International organisations funded the surveys as “service
solutionsjustasitdoessimpleoccurrencerates.This delivery surveys” and feedback of findings to spur correc-
allows for assembly of a district or national plan, made tive action was largely at CIET insistence, without funding.
up from a representative sample of local plans. A second development period focussed social audits on
methods of feedback and collation of a second order of
Methods development and lessons evidence: what communities and service workers felt
Methods developed over the 25 years fall into three gen- could be done about the problems identified in the
erations of social audit reflecting the shifts in demand household surveys. Population weighted raster maps
and supply of evidence for planning. Figure 3 portrays became integral to most social audits, sharing findings
this schematically. with planners who had limited numeracy skills or limited
In the mid-1980s in Central America, it was major pro- time to absorb findings [4,88].
gresstohavereliableevidenceonthecoverageofkey The third and current generation of social audits incor-
interventions and the indicators of their presumptive out- porates high level research methods to produce data for
comes. The first generation of social audit consequently planning, with a strong focus on analysis methods and, in
focussed on simple indicators and stakeholder discussions the area of capacity building, qualifications that could
about what could be done about them. Initial challenges develop careersfor trainees [89].Thisincludes randomised
included sampling where there was no conventional sam- controlled cluster trials in Pakistan [90], southern Africa
[91], Mexico [92], and Nicaragua [93].pling frame, designing reproducible if not standard ques-
In reaching this point, we have learnt many lessons:tionnaires, and logistics of speedy data turnaround [37].
Analysis focussed on examining associations between cov- (1) What community-based evidence to get and how
erage and impact, with sequential stratification to deal often to get it; (2) Combining qualitative and quantita-
with potential confounders and modifiers [86,87]. The aim tive evidence; (3) Moving social audit results to action;Andersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 6 of 17
Figure 3 Schematic representation of 25 years of CIET social audits.
(4) Partner buy-in; (5) Community participation; and (6) with local meaning, during the design stage. This goes
Capacity building and human capital for social audits. further than translation and back-translation, although
that happens too.
Community-based evidence: what to get, how often to Frequency of outcomes: Usually based on a cluster sam-
get it, and from where ple of households, the typical social audit is ideal for
Social audit questionnaires are ideally short, focusing on a common events. It is less useful for rarer events such as
small group of related health problems. Participation of cancer or maternal mortality. With maternal mortality
counterparts and communities in the design of question- the government priority in Nigeria, we used the cluster
naires, while desirable, can lead to longer questionnaires sample to look at common actionable risk factors for
as everyone wants to include their own concerns. Institu- maternal mortality, especially gender violence, while a
tional reviews are especially prone to collecting informa- complementary house to house enquiry laid the ground
tion that will never be analysed. It takes dedication and for measuring maternal morbidity [45].
negotiating skills to limit all instruments to items neces- Almostallsocialauditsreston voluntary disclosure,
sary to reach a decision about action. Questioning the use which filters and refracts in unpredictable ways through
of each item during design sessions is a useful filter. the gender, education, social class and culture of respon-
Standard questions and their “validation” are common dents. Rates of childhood vaccination, unofficial payments
concerns for those involved in larger scale surveys. Social and satisfaction with health services all change with type
audits have made use of standards from the earliest days of respondent. Disclosure is a real issue in sensitive topics
of indirect estimation of infant and child mortality using such as violence against women or extortion by health
the standard Brass questions [94]. We have been much workers, for fear this might result in withholding services.
less enthusiastic about using batteries of standard ques- Almost always, focus groups report higher levels of cor-
tions on culturally dependent issues, like resilience [95]. ruption than do household interviews. Health workers
We use local focus groups and several rounds of piloting themselves might be cautious about commenting nega-
to probe the local meaning of questions, and questions tively on supervisors, for fear of retribution. In PakistanAndersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 7 of 17
[51], we found that simple but standardised measures dur- Another issue is where to get the information – what
ing training of interviewers can increase disclosure about kindofsample.Samplinglies atthecore ofmethodological
violence. Interventions to decrease gender violence and rigourandthe sampleframechangedas CIETsocialaudits
corruption will change, often increase, the disclosure rate. evolved from an emergency information aid to an adjunct
This makes it harder to measure impact using a single of peacetime routine health information systems. Where
outcome indicator and underlines the need for careful the sampleframewas inadequate – andsometimes we had
an official sample frame that was simply not credible – wematching of methods of training and data collection
developedlistingsofallknowncommunities andtheiresti-between social audit cycles.
mated size. In one case we used night lights from satelliteAn important problem of a single outcome indicator is
that, while services might improve dramatically, this does pictures. In others, we used a purposive sampling method
not always produce a change in the main outcome indi- that answered the question “which 30 sites represent the
cator in the time one can allow between measurement fullrange ofconditionsacross thisregion/country?”
cycles. A partial solution is to use several intermediate Assamplingframeshaveimprovedoverthelast25years,
outcomes and to collate these with the principal outcome. wehavefound the credibilityifnotthe accuracy ofthe evi-
We have summarised these with the acronym CAS- dence increases with a random sample. Our current stan-
CADA: conscious knowledge, for example of official dard is multi-stage stratification before last stage random
costs of services or of danger signs in pregnancy; atti- cluster sampling. We stratify the sample into quadrants
tudes, that it is worth going to prenatal care; subjective (regions or provinces) then each quadrant into urban and
norms, whether most people in the area consider prena- rural. Random selection from the list in each stratum typi-
tal care worthwhile; the intention to change, to attend cally uses probability proportional to population, though
prenatal care for the next pregnancy; agency, the ability we can also oversample sub-populations as required. The
to decide to go to prenatal care or to decide where one clustersample fits withourneedforefficiencyandwe have
will deliver a child; discussion of the options is often a optimised our core analytic techniques for this approach.
precursor to behaviour change; leading to the action that A “transparency table” shows the samplecomposition next
canreasonablybeexpectedtohavethehealthimpact. towhatisknownofthepopulationproportions.
We applied this approach in immunisation [62], gender
violence [96], and HIV prevention [97]. Combining qualitative and quantitative evidence in
Overproduction of evidence: exceeding the absorptive analysis
capacity of government health services was a serious pro- Data management and analysis also evolved. We started
blem in the first years of social audits. As the evidence off using printed questionnaire sheets to help manual ana-
comes from households that cannot afford to waste their lysis in remote communities using the LT-LW model
time, this implies a serious lapse. In the Canadian Atlan- computer (Large Table-Lots of Work) available in every
tic provinces, as an extreme example, our contract with community back in the 1980s. Adhesive tape divided the
Health Canada obliged us to complete two cycles per surface into a 2x2 table or several 2x2 tables; we counted
year in each of four regions; health management systems piles of questionnaires stacked in each cell before manual
simply could not respond to this intensity of new evi- computation with the aid of a programmable calculator.
dence. We now use a two year cycle, by the end of which The arrival of laptops and software like Epi Info changed
research users are familiar with the evidence, including this, although it limited the analysis in other ways. We
the community-led solutions. went on to develop CIETmap to support our analysis
Data management methods have evolved. Our stan- approach and to interface with R, the statistical program-
dard practice is double data entry with verification of ming language.
discordant entries. In several countries we had to con- An early challenge of social audit was to include parti-
vince local statistical bureaux of the need for this by cipatory methods [98,99] in an epidemiological frame-
.demonstrating their high error rates from their usual work In the 1990s, we coined the term meso-analysis to
single data entry practice. Bhopal books have had an describe the linking of coterminous quantitative and qua-
enduring life and in some countries we still rely on litative data on groups of sites – urban/rural, or sites
these for data collection. Later social audits used scan- with particular health service characteristics [100,101].
nable self-administered or interviewer administered Our preliminary analysis simply took the site level vari-
questionnaires (we used bubbles and Remark software able into account using stratification. This evolved to
for scanning). We tried several electronic data capture include multi-level approaches, not only “taking cluster-
systems over the years. A current social audit in Nigeria ing into account” as leading to overestimated statistical
uses cellular GPS technology to geo-position the inter- confidence, but allowing that clustering is an important
view, conveying the interview in real time to a distant health development dynamic, and trying to quantify its
supervisor. effect [22].Andersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 8 of 17
Harvesting qualitative evidence has been another chal- people can join into. This does not ignore the expert
lenge. We conducted thousands of focus groups and have content of health care, or precision of technologies
to admit we have simply not exploited the full potential of involved. For example, childhood vaccination involves
the emerging evidence. Because we typically do a focus some very specific technical requirements that do not
group in each site, and a social audit may deal with hun- benefitfromparticipatory action. But people arrange
dreds of sites, it takes constant vigilance to stop people child care or transport to increase their access to vacci-
counting how many focus groups concluded X compared nation; this is a social and evidence-based process. CIET
with how many concluded Y. Several steps can make calls this socialising evidence for participatory action
interpretation of focus group discussions and key infor- (SEPA).
mant interviews less mechanical: Effective socialising of evidence at community level
Focus groups do not repeat questions in the household requires creativity to compete with the barrage of adver-
questionnaire; they comment on the results, and what can tising and the television industry, often with contradic-
be done about them. tory messages. In Mexico, social audits used song, radio
Monitors write down what people say – the words – not soap operas, community drama, comics and child-to-
counts of how many “agreed”;theynoteifaviewwas family schemes [37].
unanimous or how it was disputed; write content summa- Social audit can help to equip service workers with
ries and quotes. new tools. In Pakistan, community health workers devel-
Analysis starts by reading all focus group responses to oped training materials and communication tools using
one topic, with its prompts. evidence from a national survey on the bond of care
For each topic, monitors try to characterise the issue in between mothers and their children. The health workers
words from the group; they report this together with the embroidered the evidence on a traditional material, so
result used to spur the focus group discussion. that they could better communicate the concept of risk
Occasionally, we code a focus group outcome and use it to mothers [90]. In Afghanistan, focus groups discussed
in formal epidemiological analysis as a cluster-level vari- how to discourage people from tampering with land-
able; for example, the focus group in some communities mines (risking death or injury). They concluded that
might report “bad attitudes” or “language difficulties” of they would trust information about this from the local
health workers as a possible determinant of uptake of religious leader or from the BBC world service. Discus-
health services. sions with religious leaders led to inclusion of the issue
An advance over the last decade is our incorporation of in Friday sermons. The BBC also included the evidence
cognitive mapping to engage stakeholders in conceptualis- in their radio soap opera [1,3].
ing the focus, in design of questionnaires and to systema- In South Africa, a national youth survey on sexual vio-
tise indigenous knowledge. lence and HIV/AIDS went back to the public through
This graphic representation of knowledge of a system or an eight-episode audio-drama that presented the results
issue comprises concept nodes and causal links weighted of the survey and generated discussion that spurred peo-
according to relative importance (Figure 4). Thus ple to think about healthy sexual choices. The audio
weighted, “fuzzy” cognitive maps (FCM) offer a useful programme aired on community-based radio shows
representation of knowledge about causalities that might around the country and curriculum development specia-
otherwise seem unstructured and irreconcilable with Wes- lists made it available for life skills education curriculum
tern knowledge [102]. Fuzzy cognitive mapping is com- in schools.
monly applied as a group decision support tool to better We show elsewhere [106] how population weighted
understand complex factors contributing to a particular raster maps help to communicate evidence from social
outcome or decision [103,104]. We have used FCM to audits, especially for non-numerate audiences and set-
summarise local knowledge and beliefs around a commu- tings where broadcasting the average indicator for a sen-
nity health issue,contrasting the local beliefsystem around sitive topic is an obstacle to dissemination of evidence
diabetes to that of Western science. This expert knowl- (Figure 5). Weighted by the population represented by
edge, based on an intimate understanding of the local rea- each cluster/sentinel site, the maps the show proportion
lities, feeds into various stages of the research process, of the population affected – adding a spatial dimension
through formulating hypotheses, questionnaire develop- to this evidence.
ment, and even data analysis [105]. Recent implementation of SEPA in Nigeria began with
designation of the state-level priority – in this case,
Social audit results to action maternal mortality and morbidity. After collation of rou-
Evidence is worthless in a report left on a shelf. To have tine data on first attendances and maternal mortality
from every facility in the two states involved in the work,value, our challenge is to translate it into everyday life –
a sentinel process measuring the upstream determinantsusually meaning we must frame it as solutions thatAndersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 9 of 17
Figure 4 Fuzzy cognitive mapping. A cognitive map of prevention of HIV and sexually transmitted infections.
of maternal mortality reached out to a sample of some ayihattara). And return visits to the houses of currently
15,000 women in 180 sentinel sites who had been preg- pregnantwomenopenedadialogueaboutjustwhatit
nant in the last three years. A household enquiry docu- would take to reduce the amount of work they had to do,
mented aspects including work in pregnancy, feelings of or what would make them feel safer in their own homes.
insecurity, food security, domestic violence and access to This round of data collection focussed on those trying
care. The field teams discussed results of the household to live the solutions also reduced isolation of women,
survey (particularly issues of female genital mutilation, and it gave a message to men that someone was watch-
domestic violence and work in pregnancy) separately ing. It changed the ignorance about danger signs in
with women and men in every one of the 180 sample pregnancy and childbirth. Making the same materials
clusters. They also discussed findings with health workers available outside the sentinel sites benefits other com-
and examined health facilities serving each cluster. Ana- munities, even if they are not directly involved in the
lysis tied together this information for feedback and a social audit. This increases the impact of the social
final layer of data collection – this time about solutions audit itself, and reduces the differences between sample
and their feasibility. One page scorecards started discus- sites and other places.
sion on the issue at planning and policy level in each
local government authority (district) and at state level. A Partner buy-in, or not
15-minute video-drama told the story and raised possible A social audit ideally involves government and civil
solutions for wider discussion (www.ciet.org/Nigeria/ society, from identification of the issues and design ofAndersson BMC Health Services Research 2011, 11(Suppl 2):S1 Page 10 of 17
Figure 5 Raster map. A population weighted raster map made using CIETmap: % of respondents who do not believe that HIV infected people
must live apart from others.
the survey instruments, to analysis of the data and A social audit can sometimes be successful while docu-
implementation of communication strategies and action menting an unsuccessful programme. In South Africa’s
plans. This involvement does not always work in favour Eastern Cape Province, a social audit covered five cycles of
of quality or detail. In a social audit about childhood a regional economic development programme, the Wild
malnutrition in one country, the all male steering com- Coast Spatial Development Initiative. The development
mittee nominated by government blocked the teams initiative’s management ignored community concerns
asking mothers about their experience of violence before expressed through the social audit, and the undertaking
the birth of the child. failed [66]. Between 1998 and 2003 the Government of