"Primum non Nocere":

The traumatic consequences of "Operation Murambatsvina".

1. • That the actions of the Zimbabwe Government were precipitate, excessive and unnecessarily harsh;

2. • That very large numbers of people have been adversely affected;

3. • That very large numbers of people have been put at significantly greater risk in a variety of ways;

4. • That there is an urgent need for humanitarian assistance for those people affected by the operation.

The report of the UN Special Envoy was criticized by the Zimbabwe Government on a number of specious grounds, but, as pointed out above, the report has been corroborated in virtually every respect by local Zimbabwean research. Whilst the Government’s position on further displacements, evictions, and harassment of informal traders remains unclear, there

The Zimbabwe Government continues to deny all accusations and allegations, and shows little concern for the humanitarian consequences of this "disastrous venture", as it was termed by the UN Special Envoy. Whenever it is the subject of adverse reports, the Zimbabwe Government has shown a tendency to vilify and denigrate the authors of such reports, and then resorts to various delaying tactics in order to prevent discussion of such reports^vi. For example, the Zimbabwe Government refused to give diplomatic accreditation to a high-level official of the African Union, who had come to Harare to understand the situation and report back to the AU in its meeting in Tripoli last month.

A brief report on the preliminary findings from this most recent study was recently issued^vii, and the present report provides the detailed findings. A further series of reports may follow in future.

This report by ActionAid has been made possible through the commitment and conscientious contributions made by a number of AAI partners. The preparation of the Report would not have been possible without the support and valuable contributions of a large number of individuals and organisations. ActionAid warmly acknowledges this invaluable support and the welcoming response from all households and communities contacted during this survey. The Counselling Services Unit (CSU) provided invaluable material support and staff. The ActionAid team was also greatly assisted by the Combined Harare’s Residents Association (CHRA) and the Zimbabwe Peace Project (ZPP), who recruited enumerators at short notice to make the whole data collection process possible. The Zimbabwe Association of Doctors for Human Rights (ZADHR), and Zimbabwe Lawyers for Human Rights (ZLHR) provided valuable counsel and advice. Finally, we must also acknowledge all the households and individuals in Harare, Bulawayo, and Mutare that agreed to be interviewed, and those who provided the additional information without which this report is not possible.

A structured questionnaire was used in the collection of data from 1,195 respondents distributed in 58 affected high density wards in 3 urban centres of Zimbabwe. The questionnaire was designed to elicit detailed information on 4 thematic areas: trauma, HIV/AIDS, legal issues, and losses in the affected communities.

Data collected was entered stored and exported into Statistical Package for Social Science (SPSS) Version 13. Subsequently, analysis was done to generate frequencies, descriptive and derived variables.

The effects of trauma were assessed by means of a psychiatric screening instrument (SRQ-8), and a trauma questionnaire.

The results of the SRQ - 8 indicate an exceptionally high prevalence of psychological disorder. A total of 824 persons gave responses in the clinically significant range, 4 or more, which gave a prevalence rate of 69%. As regards an estimate of how many people will need assistance, it can be conservatively estimated that about 820,000 individuals are in need of psychological assistance, but the actual figure is likely to be higher.

Statistical analysis indicated a number of significant relationships between psychological disorder, as measured by the SRQ-8, and trauma. Our data indicated the following:

There are also interesting relations between psychological disorder and the contribution of trauma reported by time period. The strongest relationship is with trauma reported in 2005, but the trend is towards increasing levels since the 1990s.

We also examined the relationship between the frequency of trauma types over time, since it is important to understand the cumulative burden of trauma. As can be seen from the table below, there was variation between the three sites over the past few decades, and the pattern overall reflects the known history of each of these periods.

In the present study, we asked specifically about HIV/AIDS, and the data indicated that 23% of the sample was hosting at least one individual with HIV/AIDS. This was considerably higher than in the previous study – 23% as opposed to 13%. This represents a conservative number of households of about 5,407: this is an absolute minimum of 5,000 individuals whose lives are at risk. Of course, households may have more than one individual suffering from HIV/AIDS, and thus the actual number affected is much higher than this.

The effects on those with HIV/AIDS have been extremely severe. In almost every area, this sample has experienced a loss of care and treatment. The group has even lost access to nutritional support.

Our data also showed that, in every area of care and treatment, the HIV/AIDS households have seen significant and negative changes [see table over]. There is little change in access to clinics, but it must be remembered that most clinics can offer little in the way of medical treatment, except for opportunistic infections. However, very large percentages have lost access to care and treatment, with significantly high numbers receiving no care or treatment.

As regards housing, it is common for people in situations of scarce urban housing to find more informal modes of accommodation. This may lead to the development of shanty-towns, but, in Zimbabwe’s case, and based on the Zimbabwe Government’s own data, this was not a serious problem in Zimbabwe. As UNHABITAT has estimated, in 2003, Zimbabwe had a slum population of about 157,000 [3.4% of the total population], a position that was very different to the general African picture.

The effects on people’s livelihoods were equally dramatic. The greater majority reported that the Operation had affected their livelihoods, with significant percentages reporting being registered and paying rates. As can also be seen, a large number reported being assaulted whilst in police custody, and the exceeding of the statutory maximum for detention seemed common also.

It is important here to point out that we are talking about "clinically significant disorders"; that is, psychological disorders that ordinarily would require the attention of mental health professionals and that are unlikely to heal without such attention. Combined with the effects of the destruction of their homes and their livelihoods, it is even more improbable that these people will heal unaided, and there is a pressing need to develop effective psychological assistance for this population.

Our data do not allow us to specify the types of psychological disorder suffered by the population, and hence there is a pressing need for good clinical follow-up. This is imperative, since some conditions, such as depressive disorders, have higher risk than others, and the risk of suicide and para-suicide must be quickly countered, especially in the situation of high adversity that most sufferers find themselves.

Again our findings extend and amplify previous findings. As our findings show, those suffering from HIV/AIDS have lost access to quality care, medical treatment, medications, nutritional support, and many have no support at all. The HIV/AIDS group was also much more likely to be suffering from a clinically significant psychological disorder. All of this was avoidable with proper planning by the Government of Zimbabwe, but no provision has been made at all, and, worse than this, there are many reports indicating that the authorities are even obstructive to attempts to assist. This is wholly unacceptable.

We wish here to make two points. Firstly, that people’s lives are at risk is not acceptable, and it is clear that those suffering from HIV/AIDS have been placed now in a situation of multiple risks. This requires an emergency response from local and international groups, irrespective of whether the Zimbabwe Government accepts this or not.

Secondly, there must be serious concern for the effects in the long-term on morbidity due to HIV/AIDS. The risk of treatment-resistant HIV must now be seen as highly probable, and transmission a reality given the hazardous behaviours now reported in this group. The irresponsibility of a government taking such risks with public health in a situation of a massive epidemic beggars belief.

Our data do not support in any way the claims of the Zimbabwe Government that they were dealing with criminals or illegal dwellers. Rather our data suggest that these were substantial people, living with the acknowledgement and acceptance of the authorities, and conducting their livelihoods within the highly adverse economic climate of current Zimbabwe. These were not shanty-town dwellers, but people whose rights have been trampled upon. The prima facie evidence suggests that their rights, both personal and material, have been trampled upon, and who deserve redress and compensation.

The duty of care owed by a government to its citizens has been wholly violated, and it is evident that the Zimbabwe Government has failed in its Responsibility to Protect. This must require the international community to take strong and positive steps to ensure both that no further harm will come to the ordinary people of Zimbabwe, and also to provide the immediate support and assistance that these people now need desperately.

At the outset, we should state that all our previous recommendations still stand. We would wish to add a number of additional recommendations in the light of our findings:

The immediate setting up of a medical task force to provide medical care and treatment to those in need of care, with priority being given to those suffering from HIV/AIDS, women-headed households, and child-headed households;

The immediate setting up of a mental health task force to provide treatment to those in need of care, as well as the setting up of a training and supervision system for counsellors in the affected communities. Here it must be recognized that the scale of the morbidity will a long-term commitment to care, training and supervision.

The immediate setting up of a legal task force to explore the dual issues of the rights infringed and the losses consequently incurred. Since there is a very high probability that the losses experienced by the victims and the rights infringed will be excluded from legal remedy due to prescription, there is extreme urgency for this action.

Table 1: Sampled areas and proportions sampled.........................................................................12

Table 2: Prevalence of psychological disorders: Comparison of sample sites..........................15

Table 3: Comparison of scores over sample sites.........................................................................16

Table 4: Correlations between SRQ-8 and Total Trauma score per year..................................16

Table 5: Frequency of responses on Trauma Questionnaire:......................................................17

Table 6: Frequency of responses on Trauma Questionnaire [Percentages and ranks]............18

Table 7: Trauma over the years: Percentage of persons reporting trauma experience............19

Table 8: Trauma over the years: Frequency of trauma items reported.....................................19

Table 9: Effects of OM on health of households reporting member with HIV/AIDS.........21

Table 10: Frequency of psychological disorder:............................................................................21

Table 11: Care and Treatment of members with HIV/AIDS.....................................................22

Table 12: Changes in behaviour following OM.............................................................................22

Table 13: Effects on property..........................................................................................................23

Table 14 Effects on livelihoods.......................................................................................................24

Table 15: Losses of earnings & property........................................................................................24

Table 16: Estimated losses [whole sample by household]...........................................................25

AIDS Acquired Immuno Deficiency Syndrome

ARVs Anti Retro Viral drugs

HIV Human Immuno Deficiency Virus

MDC Movement for Democratic Change

NGO Non-Governmental Organization

OM Operation Murambatsvina

OVT Organized violence and torture

PTSD Post-Traumatic Stress Disorder

PLWHA People Living With HIV and AIDS

SAPP Southern Africa Partnership Programme

SPSS Statistical Package for Social Sciences

SRQ-8 Self-Reporting Questionnaire

1. - That the actions of the Zimbabwe Government were precipitate, excessive and unnecessarily harsh;

2. - That very large numbers of people have been adversely affected;

3. - That very large numbers of people have been put at significantly greater risk in a variety of ways;

4. - That there is an urgent need for humanitarian assistance for those people affected by the operation.

A total of 1,193,370 individuals were affected by Operation Murambatsvina in the six sampled urban areas. Harare was the hardest hit area accounting for 71% of individuals affected by the Operation. Bulawayo was the second worst affected nationally. However, the proportion of individuals in Bulawayo affected were less than 30% and this figure is the lowest at city or town level.

The overall population sampled was 127,587 with an average household size of 5.5. Approximately 12% of homesteads visited were headed by elders (aged over 60 years) and only 1% (142) were headed by minors (commonly referred to as child headed, below 18 years). Furthermore, children aged between 0 -17 years made up 71,691 members (or 56%) of the total population sampled. The proportion of male and female children was roughly equal (51% vs. 49%). Thirty one percent (31%) of interviewed households were hosting orphans, whilst a further 13% were hosting at least a chronically ill individual. A minority of 6% were hosting at least a mentally/physically challenged person.

Accordingly, ActionAid International, and its partners, made a number of recommendations for the support and assistance of those affected:

Regrettably, these recommendations have been almost wholly ignored by the Zimbabwe Government, as have been the recommendations of the UN Special Envoy. The Zimbabwe Government persists in claiming the necessity for the Operation, denies the adverse consequences of the Operation, and continues with forced displacements, despite stating that the Operation was complete.

The term "complex emergency" is increasingly being used to describe situations of political conflict, in the main, that result in the massive destabilization of a state’s capacity to care for its citizens^xvii. As Mollica and his associates have put this: "A complex emergency is a social catastrophe marked by the destruction of the affected population’s political, economic, socio-cultural, and health care infrastructures"^xviii. In our view, no better description could characterize Zimbabwe today, whether or not the causes are examined.

Now complex emergencies can quite clearly occur as a consequence of natural events, as in the recent Asian tsunami or the effects of Hurricane Katrina on New Orleans, but they can also occur as a consequence of human intervention as in periods of civil war or low intensity conflict characterized by organized violence and torture. We should here therefore make a distinction between, on the one hand, accidental harm causing trauma and, on the other hand, deliberate infliction of harm as is seen in wars, civil wars, low intensity conflict, genocide, and widespread political repression.

Now we are not concerned here to describe the many ways in which trauma may be inflicted during complex emergencies, but rather wish to very briefly describe their effects^xix. The

most obvious effects are physical, seen in illnesses and injuries, which may be short-lived, but also may lead to long-term disability. However, the most persistent consequences will be psychological, and especially if the trauma was deliberately inflicted^xx. Here we wish to merely emphasize three points.

Firstly, the most probable long-term consequence of experiencing organized violence and torture is the development of a psychological disorder. Secondly, the probability of psychological disorder following organized violence and torture increases with the frequency of experiencing physical harm, such as torture. Thirdly, the probability of psychological disorder increases with the number of exposures to trauma such as organized violence and torture.

Whilst there are very few good epidemiological studies of the incidence or prevalence of disorders due to trauma in Zimbabwe, there are a number of studies that are helpful in understanding the likely picture. These studies suggest that there are a number of periods in which mass trauma has occurred:

The morbidity due to the Liberation War has been the best documented to date, and the most reliable study indicated a likely prevalence of trauma sufferers of approximately 1 adult in 10 over the age of 30 years in 1997^xxii. There has been no specialized medical or psychological assistance for these victims, although war veterans have been beneficiaries of a number of occasions of compensation.

Morbidity due to the Gukurahundi has also received some attention. One small study in Gwanda District indicated that 5 adults in 10, over the age of 18 years, were suffering from significant psychological disorders, with over 90% of the sample reporting an experience with organized violence and torture^xxiii. The majority of these experiences dated from the 1980s rather than the Liberation War.

There is no reliable estimate of the morbidity due to the violence occasioned by the Food Riots in 1998. At the time the Zimbabwe Republic Police estimated that over 3,000 persons had been arrested, and the Human Rights NGO Forum was able to obtain data on 1,431 cases of persons who had been arrested. Only 44 persons eventually elected to report to the Forum, but 36% were diagnosed as having clinically significant psychological disorders. It is clearly inappropriate to extrapolate from such a small sample, but it is probable that the numbers affected were significant^xxiv.

Finally, there has been an epidemic of organized violence and torture since February 2000, and this is attested to by a vast outpouring of reports since that time. Very few studies have been done on the victims, and certainly no proper epidemiological studies. There are two indicative studies, however. The first, examining internally-displaced commercial farm workers, demonstrated that 85% of the sample were suffering clinically significant psychological disorders^xxv, whilst the second, a "snap survey" of Zimbabwean refugees in Johannesburg, Gauteng, indicated a point prevalence rate of 14% in the sample^xxvi.

As the studies demonstrate, the prevalence rates vary widely according to the population sampled. In studies of persons attending primary care facilities, estimates of morbidity are generally in the range of 20-30% of the total clinic population^xxvii. These were figures taken during the late 1980s and the 1990s, and provide a base rate for the prevalence of psychological disorder in times of "peace" and relative economic stability. Studies of the victims of organized violence and torture also vary, as we have described above, depending on the scale of the violence. Clearly, studies of populations of victims – war veterans, Mozambican refugees, or displaced commercial farm workers – show much higher rates.

With any massive displacement, and especially in the context of a complex emergency, health issues are a major concern. In the second survey, a number of issues related to health were canvassed, and these gave considerable cause for concern, as summarized below:

Clearly, the public health consequences of displacing enormous numbers of people can be extremely serious: apart from the more obvious effects around disease, in a country like Zimbabwe with a very high number of persons suffering from HIV/AIDS, an additional number of concerns that must be addressed.

When there was no apparent response from the Government, ZADHR issued a second statement making plain their concerns about the health issues:

There has been considerable acrimony over issues of legality; both the issues related to whether the Government was operating within the law during the operation, and also the issues related to the rights of persons whose goods were confiscated, or property destroyed. Many of these matters have been covered in other reports, and we will not deal with these except in passing reference^xxviii.

One area remains poorly understood however, and this is any information about the numbers of people who had valid permits for trading, or valid rights to the property that they owned. Here a number of court actions have been mounted, and, to date, most seem to have failed. However, two decisions are important, as they suggest that citizens’ rights have been disregarded.

In the first case, Justice Cheda of the Bulawayo High Court ruled that the conduct of the ZRP in destroying shacks and buildings, and in confiscating goods from vendors, was unlawful. As he commented^xxix:

Useful as the small number of court actions are in understanding the limits of legality for both Government and citizens, they cannot give any impression of scale. As was noted in the second ActionAid International report, very high numbers reported losses:

The pressing issue, and one related to some of the conclusions in the report of the UN Special Envoy, is the number within this overall total that had valid rights to practice their business or occupy the property that they lived or worked in. This clearly deserved further investigation.

Assuming on the basis of the ActionAid data, that Harare was 60% of the national loss, then this figure would look more like Z$1,221,335,000.00 [US$73,574,397.00]. Of course, this only deals with direct losses, and does not include the potential losses of earnings from rents, selling, small scale manufacturing, services, etc.

Additionally, we should recall the comments of the recent Article IV consultation by the International Monetary Fund. Noting the massive decline in the Zimbabwe economy, the IMF commented:

Clearly the IMF has expressed considerable concern over the plight of ordinary Zimbabweans, and, in the assessment of the Executive Board of the IMF, they commented further:

Overall, the IMF report indicates the need for wide-reaching reform to be undertaken by the Zimbabwe Government in order to stop the economic decline, and ameliorate the consequences for the very poor.

Given the dearth in all these estimates, and the pessimistic views over the likely performance of the Zimbabwe economy, it is clearly important for ActionAid International to understand how Operation Murambatsvina has impeded the ability of ordinary Zimbabweans to cushion them against this exceedingly harsh economic climate.

A structured questionnaire was used in the collection of data from 1195 respondents in 3 urban centres of Zimbabwe, namely Harare, Bulawayo, and Mutare.

Administrative boundaries were used since these would help in the identification and prioritization of wards and assess the impact of the Operation and direct influence on defining livelihood characteristics of households. Furthermore, this ensured that overlapping of areas covered was avoided. Three urban areas were selected based on the previous survey, which suggested these were the worst affected areas. An enumerator was responsible for data collection in each ward over a 3 day period. A total of 58 wards were surveyed, as follows:

Harare is the capital city of Zimbabwe with an estimated population at 1,444,534 based on the most recent census (excluding Chitungwiza which is home to 321,782 and Epworth, 113,884). Most industrial and commercial companies are located in Harare. Harare has 45 administrative wards and of these 30 are high density suburbs. Offices of government are located in Harare. Currently city council affairs of the city are being run by a government appointed commission after the line minister responsible for local government fired the elected opposition MDC mayor and the 45 ward councillors.

Bulawayo is the second largest city in located in the south western part of Zimbabwe. Bulawayo has an estimated population of 676,787 people. Until recently Bulawayo had a thriving industrial sector however, due to perennial water shortages and the general decline in the economy, Bulawayo’s industrial base has been severely compromised. Bulawayo has 29 administrative wards and of these 19 are high density suburbs. The incumbent mayor for Bulawayo is from MDC.

Mutare is located in the Eastern highlands of Zimbabwe, a scenic tourism area, and shares the border with Mozambique. It is the third largest city in Zimbabwe, after Harare and Bulawayo, with a population of 170,106 in its 18 wards, of which 12 wards are high density suburbs. Currently, the city council is headed a government appointed commission. The elected MDC mayor was relieved of his job after accusations of mismanagement by the line ministry. However, speculation is rife that this is in retaliation after the mayor facilitated the UN Special Envoy’s visit around Mutare after the commencement of her fact finding mission.

The Self-Reporting Questionnaire [SRQ-8] is a widely-used psychiatric screening instrument, developed in Zimbabwe. It was derived from the Self-Reporting Questionnaire [SRQ-20], developed by the World Health Organization in 1980, and the SRQ-20 has been widely used in Africa as well as in other developing countries^xxxiii. The SRQ-20 was developed in order to provide an instrument for reliably detecting non-psychotic mental disorder, and has been shown to be both valid and reliable in a large number of different African settings, including Zimbabwe^xxxiv. The SRQ-20 has additionally been used in Zimbabwe in a number of studies of the effects of organized violence and torture^xxxv.

The SRQ-8 was derived from the SRQ-20, and shown in a Zimbabwean study to be both reliable and valid^xxxvi. It is a simple 8 item questionnaire, investigating 8 common symptoms in the past week. All scores in excess of 4 can be taken as indicating significant psychological disorder.

The Trauma Questionnaire was based on the Harvard Trauma Questionnaire [Vietnamese Version] that had been modified for use in previous Zimbabwean studies^xxxvii. The Harvard Trauma Questionnaire has been widely used in a variety of different settings, and a number of versions have been developed for specific settings^xxxviii.

The Harvard Trauma Questionnaire is usually given as a screening instrument for the detection of disorders due to trauma, but in the present study this was done by means of the SRQ-8, with the Trauma Questionnaire being used purely as a method of reporting upon the kinds of trauma experienced and witnessed by the interviewees.

Previous work conducted which included collection of qualitative data in previous studies and emerging issues guided the structuring of the instrument. Further, issues raised in the Special Envoys Report also helped in the formulation of questions pertaining to HIV and AIDS. Of special interest was quality care (Home Based Care) and comprehensive care (ART and treatment to opportunistic infections). As already highlighted people living with HIV and AIDS are a special interest group since their needs are critical and due to the high HIV and AIDS prevalence rates in Zimbabwe.

Questions surrounding the legality of Operation Murambatsvina have been a point of contention, with the GoZ claiming that it was acting within the laws of the land. However, civic society strongly disagrees. A lot has been said on who had and had no approval on trading and construction of homes. Therefore as a follow up to this, questions were designed to capture these aspects.

The impact of Operation Murambatsvina has been debated both by civic society, actor alike and GoZ, with the later claiming minimal losses. In the national study conducted by ActionAid losses recorded were substantial. However, this was not disaggregated data. Losses were recorded as losses. In this study, an attempt is made to differentiate income and property lost as a result of the Operation.

Initially, after data collection, questionnaires were screened and post-coding done. Thereafter, information collected was entered into a central data base. Subsequently, data was cleaned, processed and analyzed in using both Statistical Package for Social Sciences (SPSS) Version 13 and Microsoft Excel. In addition, a discussion was held with enumerators to gain additional insight and emerging issues that was not captured on the questionnaire.

In the following sections, we present the findings related to trauma prior independence to date.

The results of the SRQ-8 indicate an exceptionally high prevalence of psychological disorder. A total of 824 persons gave responses in the clinically significant range, 4 or more, which gave a prevalence rate of 69%. The mean score for the whole sample [n=1195] was 4.8[sd.2.5]. This is nearly 3 times higher than previous rates obtained from a general setting in Zimbabwe, and is much more in keeping with the rates obtained from populations with obvious trauma, such as refugees or victims of war.

As regards an estimate of how many people will need assistance, it can be conservatively estimated that about 820,000 individuals are in need of psychological assistance, but the actual figure is likely to be higher, as we only interviewed one person per household, and whole households are likely to have been as affected as the persons interviewed.

As can be seen in Table 3 above, there are regional variations in the scores derived from the questionnaires, but, overall, there are significant relationships between the probability of psychological disorder and the total number of trauma events reported [Total Trauma Score], the number of trauma events indicative of violence [OVT items], the number of trauma events indicative of displacement [OM items], and the number of different time periods in which trauma was experienced. As can be seen [footnote 24 below] the strongest relationship was with OM items on the Trauma Questionnaire.

Therefore, our data show the following: