Final Report to Unaids: Police and Sex Workers in Papua New Guinea

Summary

Introduction

As approved in the original proposal, the specific objectives of this intervention have been to:

• increase the frequency of condom use among sex workers and their clients
• increase their level of knowledge and decision-making competency with regard to disease prevention and overall risk taking in sex and the sex trade
• diminish selected factors (e.g. sex with drugs and alcohol, low self-esteem, fear of police, fear of STD clinics, punitive attitudes of police and health care workers) which create barriers to preventive behaviours among sex workers and their clients
• encourage normative change toward safer sex practices among men who are frequently clients of sex workers as well as those in the periphery of the sex trade
• evaluate, by both qualitative and quantitative means, the expected changes in knowledge, attitudes and behaviour
• develop the capacity of AIDS prevention workers to carry out and evaluate behavioural change projects within the context of an active NGO

Monitoring and evaluation

Baseline surveys were conducted with sex workers and their most high-risk client groups. As recruitment and training of personnel had to take place as the first priority and action had to start quickly, the timing of baseline surveys was staggered to take place before substantial intervention with each group began. Interviews were standardized after considerable pre-testing. Pre-testing of baseline interviews with sex workers actually began in 1995 before the project funding was approved, in anticipation of the coming intervention. Some client groups, such as sailors, are highly mobile and the analysis will combine those interviewed in both Lae and Port Moresby. Baseline questionnaire forms are attached. The following schedule of baseline and follow-up surveys have been accomplished.

Sample	Location	Date Start	Date End	Number Interviewed
sex workers	Port Moresby (streets)	1995 Aug	1995 Oct	phase 1 - 79
pimps	Port Moresby	1995 Aug	1995 Oct	20
clients (mixed)	Port Moresby	1995 Aug	1995 Oct	68
sex workers	Port Moresby	1996 Mar	1996 Sept	phase 1 thru Mar; phase 2 - 297*
dock workers	Port Moresby	1996 Mar	1996 June	84
dock workers	Lae	1996 July, 1997 Jan	1996 Oct, 1997 Feb	97
sailors	Port Moresby	1996 Mar	1996 June	197
sailors	Lae	1996 July, Sept	1997 Jan	54
policemen	Port Moresby	1996 Sept	1996 Nov	phase 1 - 130
policemen	Port Moresby	1997 May	1997 June	phase 2 - 197*
police women	Port Moresby	1996 Aug	1996 Sept	53
security men	Port Moresby	1996 May	1996 Sept	154
sex workers	Lae	1997 Jan	1997 Feb	181
TOTAL				1,611

Due to time constraints and shortage of staff to accomplish all the tasks required, only two groups have had any follow-up survey work. These are Port Moresby sex workers and policemen, and are marked with an asterisk in the above table. The first group to receive attention from the project was the sex workers, in March 1996. During this time, negotiations began with police officials to inform them of the project's intentions and concerns. While cooperative attitudes were noted immediately, official permission had to be granted by the Police Commissioner. After several months, a letter was received and the baseline surveys began.

Increasing the frequency of condom use among sex workers and their clients

Peer education

Peer educators were trained in all target groups. A set of four modules was developed for training sex worker peer educators . Because these are all floating or street sex workers, they have no natural groups but keep moving around the city's venues on an individual basis to maximise their access to clients. Consequently, they can be brought in for training only in small groups. As of September 1997, 28 sex workers had completed all four modules, 37 had completed three, 62 had completed 2 and 137 had completed one, for a total of 264 women trained, at least, in part. In order to accomplish this, 78 sessions had to be held.

In August the police sent one of their welfare officers to attend the week-long training of peer educator trainers. After that, police peer educators were selected by the police themselves and began training. To date, a total of 32 policemen peer educators, representing 13 police stations, and 16 peer educators among police women (at 10 police stations) have completed training. These persons have been followed-up by the outreach workers to support and supervise their work. There have been up to 6 visits to some of the policemen peer educators, and up to five visits to policewomen. In total, 101 follow-up visits to policemen and 40 visits to policewomen peer educators have taken place. The policemen peer educators have organized workshops at police stations and headquarters. The men have held 10 and the women have held 6 workshops. Although no baselines were ever obtained from policemen's wives, under their own pressure, 10 peer educators among them have been trained. All police stations and barracks in Port Moresby have been covered by this programme. Thus, a large number of police and their families have been reached through this process.

Condom distribution and use

The required government condom supply is greatly underestimated by those who order them. This situation may have been ameliorated had the social marketing of condoms component of the AusAID HIV and Sexual Health Project been successful. Because it has barely been developed, no inexpensive condom is widely available on the market. The least expensive is the brand that had been socially marketed earlier in 1991 by SOMARC, Protector condoms, which sell at K.75 for 3, i.e. at a price equivalent to approximately US10 cents per piece, at some urban shops. When condoms are unavailable, outreach workers advise sex workers and others to purchase these. However, they are not widely distributed and are not accessible where most needed. Currently, UNFPA has purchased 15,000 female condoms and has promised to give a large number to the project. These have been in the country for 5 months but are being held up by a debate over an accounting form. AusAID states again that another container load of male condoms is due soon.

In total, the project has distributed 845 female condoms and 68,200 male condoms to sex workers. Policemen and women in Port Moresby have received at least 188,000 condoms to date. The police distribute them to many men who they think are likely to have multiple partners, not just to police. Some also report that they give out strips of five to anyone coming to the police station with a complaint of any nature. Hence, they have become a major distribution force for the city. Security men have received only about 10,000 to date.

Evidence of the increasing use of condoms is found in the follow-up survey data, as well as anecdotal reports. Among the sex workers, in phase 1: Aug 95-March 96, a sample of 129 women with 347 clients had 693 acts, for which no condoms were used on 528 of those acts. This calculates to a use rate of 24%. By phase 2: Mar 96-June 1, in Port Moresby, after our workers had begun their outreach, the women had had 539 partners and 576 sex acts, during which condoms were used 253 times for a use rate of 44%. This represents a significant difference (p=.000, M-H chisq=138.27, Yates corrected chisq=137.00, 95% C.I.). Between June and October, when the survey was completed, there were no further significant increases (had risen to 49%).

These data also indicate an increase in condom use with non-clients. In phase 1, of the sample of 129 women, 83 had had sex with men other than clients, i.e. husbands, boyfriends, casual partners, for a total of 99 bouts of sexual intercourse. Of those 99 bouts, for 89 no condoms were used, making a use rate of 10%. By phase 2, out of 297 women, 158 had had sex the previous week with 190 men. Of a total of 378 bouts of intercourse, 258 had not had condom coverage, making a use rate of 32%. This increase also calculates as a statistically significant difference (p=.000, M-H chi sq=18.5, Yates corrected chi sq=17.46, 95% C.I.).

Among police condom use has also risen, particularly with commercial and casual partners. In the pre-intervention phase, out of a sample of 130 men, 33 had had sex during the last week with 57 women for 76 bouts of intercourse. The number of bouts of intercourse per woman ranged from 1 to 11. 39 of these bouts had not been with condoms, making a use rate of 49%. After the intervention had progressed and a re-survey conducted, of a sample of 197 men, 35 had had sex with 45 casual partners for 77 bouts of intercourse (range 1-10). Of these, 23 bouts had no condom coverage, making a use rate of 70%. Testing for significance, p=.006, uncorrected chisq 7.3;p=.007, M-H chisq 7.25; p=.011, Yates corrected chi sq 6.44). It also appears that the proportion of men seeking sex with 'rot meri' (women of the streets) has decreased from 25% to 18% between the two samples. This may not be significant and was never an objective of the project. Future re-surveys will be necessary to examine the reality of this trend.

With their wives and steady girlfriends, condom use did not rise significantly. In phase 1, 84 men (of 130) had had sex the previous week with 85 women for 229 bouts of intercourse, 206 of which had not been covered by condoms. In phase 2, 123 men (of 197) had had sex with 134 women, for 310 bouts of intercourse, 272 of which had had no condom coverage. This comes to a use rate of 12% (not significant, p=.5, Yates correction). One of the issues that arose in discussions with policemen's wives was that they would be insulted and angry if they saw condoms carried home by their husbands. Clearly, a great deal more work must be targeted at policemen's wives and, with policemen, some specific work concerning the options for safer sex within the family (or in important relationships).

Among security men, no follow-up has taken place. Baseline survey data indicated the following: of a sample of 154, men, 55 (36%) had had sex the previous week with 79 commercial sex workers for a total of 150 bouts of intercourse. Of these, 114 had not been covered by condoms, making a use rate of 24%. In addition, 85 men had had sex with 88 girlfriends and wives for a total of 252 bouts of intercourse, of which 246 had not been covered by condoms. The use rate with 'significant others' amounted to 2.4%. Given the high level of commercial sex, these men and their wives and girlfriends are at a high risk of acquiring HIV and a great deal more work must be accomplished with them.

Increasing knowledge and decision-making for safer sex

However, among police the pre- and post-test sample sizes are adequate. These indicated very low levels of knowledge of non-sexual (actually non-heterosexual) means of HIV transmission both before and after the intervention. Information levels had changed, however, on STDs. Knowledge of the fact that some STDs may have no symptoms rose from 19% to 53%. Knowledge of ways to avoid getting HIV/STDs also rose, with 'always using a condom' rising from 49% to 64%. Levels of correct risk perception also changed among policemen. Among those considering themselves to be at high risk, the percentage of correct reasons rose from 33% to 75%. Among those who classified themselves as low risk, the percentage of correct reasons rose from 33% to 51%.

Sources of information also shifted. Asked if anyone had come to them to discuss HIV/AIDS, the proportion answering positively rose from 8% to 38%. When asked which agency they had learned about AIDS from, in the first survey the majority (93%) mentioned the PNGIMR. During the second survey, the proportion mentioning the Papua New Guinea Institute of Medical Research (PNGIMR) dropped to 84%, while the police peer educators rose to 13%. All other sources of information on the topic (TV, radio, newspapers, etc) dropped over the 9 months of measured intervention.

Diminishing selected factors that create barriers to safer sex

Power structures and gatekeepers

Security firms were more difficult to convince of the importance of AIDS prevention than were police, despite the alleged deaths of several security men from AIDS. Security firms occupy a large wage labour sector in Papua New Guinea. The project has contacted a total of 32 such firms in Port Moresby. The actual number existing is unknown as many are not legally registered. After many months of rejection (even complaints about being approached by the project sent to the Director of PNGIMR), a few security firms became cooperative. They were then asked how we could convince those who reject our advances, which included the largest in the nation. They suggested that a workshop be held for security firms and be televised. In July, 1997, such an event occurred, with cooperation from the DOH, at which 27 security firms were represented. The TV workshop was quite successful.

Unfortunately, the number of outreach workers in the Port Moresby office (as in Lae) are stretched very thin. As the number of reported cases of HIV rises dramatically in PNG, visible concern has also risen and many meetings and workshops are called, taking up the time of the project personnel. While there are plans of training peer educators among the security firms and the firm managers have selected men for the purpose, it is considered necessary to interview each one before beginning. So far, 26 firms have submitted names of 42 prospective peer educators. For these reasons, the training of peer educators among security men has been slow to start, but will take place shortly.

Sex workers in Papua New Guinea have far fewer men (or women) who control their lives and incomes than do sex workers in Asia, for example. The chief gatekeepers are the owners of the venues where the workers gain access to clients and where they and their clients go to have sex. These are often older men who collect rent from them for rooms. In several cases, the project has made direct contact with these men and distributed condoms to them. Some, unfortunately, were found to be selling the condoms at exorbitant prices to the sex workers or their clients. This has been stopped. In some cases, these 'guest house' owners have adopted a 100% condom policy. Signs in Tokpisin have been placed in every room stating that only sex with condoms is allowed. If the owner finds out that condoms were not used, he will not allow the sex worker or the client to use his premises again. One owner stated that he liked this policy because, if one of the clients or sex workers got HIV, he could not be blamed. In the larger hotels, condom machines have been installed in the men's rooms and now the major hotel chain has developed a courtesy package cover, to be filled with condoms supplied by the project and with the hotel's name on it, to be placed in every room in the three hotels owned in Port Moresby. This chain also owns hotels in other cities and towns, therefore, this innovation has the potential for much wider spread.

Over the course of this intervention, police have raided some of these 'guest houses', and, in one instance in 1996, jailed 19 sex workers, of whom several were peer educators. This created a terrible delay in the project as the women thought they had been arrested due to their association with the project, and subsequently vanished. It took several months and a number of discussions with the police to bring this situation under control.

Lainaps (line-ups)

This practice was strongly targeted by a number of strategies. Diagrams and demonstrations were used during the peer education training to show the men how rapidly HIV can spread through such group sex events. A comic book specifically about lainaps, in which an HIV negative sex worker is forced into sex with numerous police, who later are shown to be HIV positive (see attached). The story was designed in order not to place blame on sex workers but to illustrate the possibility of acquiring HIV from the semen of previous men in the line or queue. The flip chart given to each peer educator also has a specific drawing illustrating the practice, though not among police, in order to bring it to immediate attention when discussing risk behaviours in the PNG context.

In sum, it appears these efforts have begun to bear fruit, at least among the police. In the pre-intervention phase, 10% of the men (n=130) stated they had been in a line-up during the previous week, whereas this was reduced to 4.2% post-intervention (n=197). The average number of men involved has not changed significantly, 4.22 vs 3.58. Among the security men, 11% of a sample of 154 stated they had been in lainaps the previous week, with an average of 5.29 men per event. As the security intervention develops, it will be important to target this highly risky practice among these men also.

Enabling approaches: raising self-esteem

It appears that some sex workers find that their level of knowledge about HIV is far greater than that of the surrounding communities in which they live. As sex workers in PNG are far less stigmatised than in many nations, they are respected for this knowledge and requested by the youth group, women's groups and others in their communities to give talks about HIV. The video collection at the project house has initiated a lending system for sex workers, who use them in educating the communities in which they live.

While it has always been difficult for a sex worker to gain redress for rape or theft, it has been nearly impossible for her to prosecute a policeman. There are now some indications of change. In August 1997, one of the 'guest houses' that is a regular part of our project outreach, was raided by 6 policemen who took advantage of the situation to rape two sex workers. The sex workers were arrested but they managed to lay complaints against the police officers. All 6 policemen were jailed, pending court, and the sex workers were set free. A newspaper article chronicling this event is attached. A few months before this event, a non-sex worker woman laid complaint against being raped by the police. It appears that increasingly more women may become willing to stand up for their right not to be raped by police, a positive sign for progress in both HIV prevention and human rights.

Facilitating STD treatment

Table 1. Confirmed incidence of HIV infected persons per year in PNG

1997 181 until end July, end year projected (PNG HIV National Reference Laboratory).

Given their increasing experience with AIDS, sex workers in Port Moresby have generally become more amenable to perceiving the need of safe sex and STD treatment. During phase 1, 34 out of 87 (39%) women stated they currently had symptoms of STDs, whereas during phase 2, 40 out of 297 women (13%) stated they had current symptoms. Over that period, women had been shown STD slides, two comic books about STDs and sex workers were distributed, peer educators were given STD photo albums to use for education, and a special entrance and day at the government STD clinic had been set up for sex workers. Project personnel accompanied these women to the clinic and provided transport. Such a reduction, if real, needs to be validated with clinic data.

Improving knowledge and practice of safer sex

Gradually, new ideas in sex work are spreading. Condom skills, in particular, have been emphasised. Sex workers are asked to practice putting a condom on (a phallic model) in the dark or blindfolded and have been shown the technique of putting a condom on with the mouth.

Encouraging normative change among clients of sex workers towards safer sex

The major clients of sex workers with whom this project works are sailors, dockworkers, police (not really clients in that they seldom actually pay for sex), security men and, in Lae, truckers. There are other important groups, such as civil servants, soldiers, mine workers, construction workers, and loggers. These have not been targeted in any way by this project. Among all of the men in the transport industry, however, many peer educators have been trained, materials and condoms distributed and outreach work continued. A considerable number of the clients of sex workers are men from the slum communities in which they live. Recently, the sex workers have initiated HIV education for some of these slum (called 'settlements' in Port Moresby) communities.

One other method of reaching clients as well as the general public that has begun is a telephone hot line. This telephone line was activated in June 1997, but remains to be advertised and therefore, has not been well utilised yet. One issue that slowed down its use was that no one was trained to do telephone counselling. By early September, 2 of the project's outreach workers had completed HIV counselling courses and several more are scheduled to undergo training in October. T-shirts advertising the hotline number are under production.

A newsletter to be sent to all the peer educators has been suggested, mainly by the sailors. This innovation is under consideration and may be useful for reinforcing messages among those who are literate. This may not be too useful for many sex workers directly, but could impact on their lives through better education for their clients.

Developing the capacity of AIDS prevention workers to carry out and evaluate behavioural change projects within the context of an active NGO

The Project's Future

With the completion of this funding period, it is clear that considerably more work must be accomplished. Additional funds will be sought for further work with security firms, army and other known groups of men who are frequent clients of sex workers. It is expected that AusAID funding will continue, although a gap may occur during 1998-1999. New sources of funds will also be sought to continue the growth of this project into other sectors of the population. It is obvious, however, that foreign donors cannot be depended upon forever and that new habits (eg. buying condoms, decreasing numbers of sexual partners) and new facilities (eg. friendly, accessible and correct STD clinics) must become part of a new social order in Papua New Guinea, either by way of government or the NGO sector.