The information below was derived mainly from structured observations and from notes taken by researchers who visited the facilities.

Structural issues

Both health care providers and ARV users highlighted a number of structural problems in the health facilities which had a potential impact on adherence to ART. At Mount Meru in Arusha, for example, there was no separate room for consultation and thus no possibility of confidentiality for patients. At the time of this survey, three doctors were sharing a single room and consulting with three different ARV users at the same time. However, ARV users frequently mentioned that they were accorded respect. Elsewhere, at St Elizabeth, the waiting area was limited. In Arusha, the faith-based facilities generally had better operating structures than those in the public sector, where confidentiality was compromised by lack of space and the number of consultation rooms that were shared.

In contrast, consultations in Dar es Salaam took place in more appropriate consultation rooms. Although the quality of the infrastructure varied between the different facilities, it provided for adequate confidentiality, good counselling and adequate laboratory services.

The public health facilities in Dar es Salaam are well-endowed due to their involvement in the MDH project and operate on a daily basis unlike most of the facilities in the Arusha region. For instance, of the two public sector hospitals studied in Arusha, Mt. Meru had only one ART clinic per week, while Arumeru was running ART clinics daily (Monday to Friday), and the faith-based facilities were operating ART clinics three days per week.

Other structural problems highlighted included: the lack of prescribing capacity at Arumeru Hospital, where ARV users went home without medication on days when the hospital's only prescriber was not on duty; limited waiting space at the Hindu Mandal private facility in Dar es Salaam, which was difficult for both ARV users and care providers; lack of transport for the staff involved in home-based care services; and lack of medicines at ART clinics for the treatment of HIV-related opportunistic infections.

In addition, a female FGD participant in Arusha complained about the rigid bureaucratic procedures involved in transferring to another ART clinic closer to home:

"I was registered to start ART clinic in Kilimanjaro Christian Medical Centre (KCMC) in Moshi a year ago. At that time there was no ART clinic near my village. Now there is a clinic near my home but I am denied transfer from KCMC to my home clinic. KCMC is very far from here, about 170 km away. Sometimes I do not have the fare to travel to KCMC, hence missing my doses."

Key informants also maintained that there were not enough health care facilities providing ART. Meanwhile, ARV users complained that services in the few existing facilities were deteriorating due to the increasing number of patients. One key informant from Kinondoni, Dar es Salaam, suggested that:

"ART clinics should be increased and more private hospitals should also be included in the programme."

There were reports of wide variations in the length of time ARV users had to wait at the clinics. According to ARV users, waiting times varied from less than one hour to 10 hours (see Table 8 below).

Table 8: Average time spent by ARV users in health facilities studied

SSI: Semi-structured interview

Staff qualifications and working conditions

A total of 28 staff were interviewed (including 16 from the four nongovernmental facilities). They included a dietician, laboratory technicians, counsellors, medical doctors, pharmacists and social workers. Only one of them had primary school education only and five had studied up to secondary education level only. The remaining 22 (78%) had undergone various tertiary education, including college education.

On the day of the exit interview, 97% of all patients saw the doctor and 77% saw a pharmacist while only 21% saw a counsellor, as presented in Table 9.

Table 9: Categories of health staff seen by patients on day of exit interview (N=70)

The staff interviewed had worked in ART clinics for a period ranging from 3 to 36 months, with an average of approximately 15 months. Some staff (21.4%) had 12 months' experience of working in ART clinics. However, only three (10.7%) of the respondents reported that the training they had received for ARV management was adequate. Twenty respondents (70.7%) said that the training they received was too short and that they needed additional training.

A major complaint was the pressure of work. A nurse in one facility in Dar es Salaam said:

"There is a lot of work in this ART clinic and we are overworked."

As a result, many of the ART staff had multi-functional roles. For instance, one counsellor was responsible for routine nursing duties as well as administrative and supervisory duties. Out of 28 respondents, 11 (39.3%) were involved in adherence counselling in addition to other routine duties. In addition, nine nurse-counsellors were also dispensing drugs.

All the health facilities surveyed indicated that they were receiving an increasing number of ARV users, thereby adding to their workload. A female participant of an FGD in Mwananyamala said:

"If the situation remains like this, doctors will be tired and the last patient will not be attended properly."

Meanwhile, a staff member from a hospital in Arusha suggested that doctors should start the clinic earlier.

In Dar es Salaam (but not in Arusha), the staff in public facilities are provided with some topping-up allowances. However, the majority of health care workers (93%) expressed low motivation. Other challenges are shown in Table 10 below.

Table 10: Challenges most frequently mentioned by health staff (N= 28)

In summary, staff complained of the increased workload involved in treating ARV patients, for which they were not adequately trained. In contrast, patients largely expressed appreciation for the quality of care provided, despite complaints about long waiting times.

Availability of guidelines and diagnostic equipment

A number of facilities lacked the necessary laboratory and diagnostic equipment. This led to delays for patients as they had to return again in order to get the results of tests (see Table 11).

Table 11: Status of diagnostic facilities in health facilities studied

The table shows that of the seven health facilities, only three had both CD4 and biochemical testing machines. According to both ARV users and staff, the services of the laboratory and testing were inadequate in some of the surveyed facilities. Problems cited included the unavailability of CD4 reagents and of CD4 cell count machines. As a result, test results were delayed because they had to be carried out in other facilities elsewhere.

All the facilities had NACP guidelines for ART. In addition, one nongovernmental facility in Dar es Salaam had WHO guidelines as well. The MDH project had also provided guidelines to its sites in Dar es Salaam. However, none of the facilities visited had their own policy guidelines.

Staff from all seven facilities reported that they had adequate HIV testing facilities and reagents and had no problem with processing results. However, two facilities in Dar es Salaam were concerned about delays in obtaining results.

Availability of medicines

The prescribed ARV drugs were usually available in all seven facilities. Of the interviewed staff, only five (including four from government facilities) reported periodic shortages of prescribed medicines, requiring them to borrow medicines from nearby facilities. However, it was learned retrospectively during the course of this study that ARV drugs were out of stock for a whole month in Arumeru Hospital, causing an interruption in treatment for ARV users. The unavailability of ARVs was linked to a delay in supplies from the Medical Stores Department. However, this situation has now been rectified. About 36 patients were on ARVs at that time and were asked to get their supply from the nearby Mt. Meru Regional Hospital.

During the survey period, there was also a shortage of supplies of Triomune 40 at the Mt. Meru Hospital. When needed for dispensing during the clinic day, the hospital had to borrow supplies from Arumeru Hospital.

Although medicines for opportunistic infections were reported to be available in some of the facilities, these were not provided free of charge in faith-based and private facilities with the exception of anti-tuberculosis medicines. In Arumeru Hospital in Arusha, medicines such as fluconazole and other antifungal medicines were not available, and there was no substitute drug for clients who had experienced an adverse drug reaction after using cotrimoxazole.

Since all facilities providing ART were getting medicine supplies from the Medical Stores Department, there were no significant variations in the available stock of ARVs. However, Hindu Mandal and Selian, both faith-based facilities, showed some slight variations in stocked medicines. For example, Hindu Mandal stocked indinavir and didanosine and Selian stocked abacavir and didanosine, medicines that were not available in other surveyed facilities. In addition, there were no fixed-dose combinations for syrup preparations, which poses a problem for paediatric treatment. In some of the facilities surveyed, there was inadequate space for drug storage and for counselling during the dispensing of ARVs.

Provider-patient interaction regarding use of ARVs

In all the ART facilities studied (both public and private) respondents stated that providers showed respect for ARV users. This may be the result of initial training, which was conducted by NACP under the MoH. One male FGD participant said:

"Providers show a warm relationship to clients and provide good services." (Male FGD, Mwananyamala, Dar es Salaam)

However, hospital support staff who had not received any training on how to deal with AIDS patients were reported to be stigmatizing ARV users and behaving in an unethical manner towards them. For example, it was reported that a hospital cleaner in one of the public facilities in Dar es Salaam had shouted loudly to indicate the place where ARV users were seated, waiting to be attended to. This had been a major source of embarrassment for all involved.

As part of the follow-up on treatment adherence, health staff were expected to ask specific questions on the use of ARVs. Figure 4 below summarizes the questions most frequently put to ARV users. These data were obtained during the observation of consultation phase of the study.

Figure 4: Questions most frequently asked by health workers during follow-up in ART clinics (N = 64)

During observation of consultations, seven new ARV users who came to start treatment were told about the possible side-effects, six were told about the importance of continuing with treatment, five discussed how to use ARVs, and four were informed about what to do if they forgot to take a dose. However, there was no discussion about other reproductive health needs, such as contraception and safer sex, in any of the observed consultations.

Costs

The cost of travel to the health facility for ARV users ranged between Tshs 200 to Tshs 30 000 (approximately 20 US cents to US$30.00), while the distance travelled ranged from 1 km to 246 km. Most patients had started ART less than one year ago (mean 8.7 months, range 3-36 months). Costs were described as a constraint to adherence. Some ARV users said they were unable to keep their appointments because they could not afford the cost of travel. Patients also cited other costs, including registration and consultation fees. A registration fee of Tshs 1500 (US$ 1.50) was levied at all facilities in Arusha except Mount Meru, where the registration fee was Tshs 3000 (US$ 3.00). Elsewhere at Selian Hospital in Arusha, in addition to the registration fee, patients were also charged a consultation fee of Tshs 3000 (US$ 3.00) per visit and Tshs 15 000 (US$ 15.00) for each laboratory investigation to monitor the CD4 cell count. With the exception of the private Hindu Mandal Hospital, which charged a monthly consultation fee of Tshs 1000 (US$ 1.00), the facilities surveyed in Dar es Salaam were not charging any fees.