PUBLIC HEALTH PRACTICE
Evaluación de servicios de asistencia ambulatoria en sida, Brasil: estudio comparativo 2001/2007
Maria Ines Battistella NemesI; Tatianna Meirelles Dantas AlencarII; Cáritas Relva BassoI, III; Elen Rose Lodeiro CastanheiraIV; Regina MelchiorV; Maria Teresa Seabra Soares de Britto e AlvesVI; Joselita Maria Magalhães CaracioloI, III; Maria Altenfelder SantosVII
IDepartamento de Medicina Preventiva. Faculdade de Medicina. Universidade de São Paulo. São Paulo, SP, Brasil
IIDepartamento de DST, Aids e Hepatites Virais. Secretaria de Vigilância em Saúde. Ministério da Saúde. Brasília, DF, Brasil
IIICentro de Referência e Treinamento em DST/Aids. Secretaria Estadual de Saúde. São Paulo, SP, Brasil
IVDepartamento de Saúde Pública. Faculdade de Medicina de Botucatu. Universidade Estadual Paulista Júlio de Mesquita Filho. Botucatu, SP, Brasil
VPrograma de Pós-Graduação em Saúde Coletiva. Departamento de Saúde Coletiva. Universidade Estadual de Londrina. Londrina, PR, Brasil
VIPrograma de Pós-Graduação em Saúde Coletiva. Centro de Ciências Biológicas e da Saúde. Universidade Federal do Maranhão. São Luís, MA, Brasil
VIIPrograma de Pós-Graduação em Ciências Área Medicina Preventiva. Faculdade de Medicina. Universidade de São Paulo. São Paulo, SP, Brasil
OBJECTIVE: To assess Brazilian Unified Health System outpatient services delivering care to adults living with AIDS in 2007 and to compare with the assessment conducted in 2001.
METHODS: The 636 health services registered in the Ministry of Health in 2007 were invited to respond to a previously validated questionnaire (Qualiaids Questionnaire) with 107 multiple-choice questions about the organization of care delivery. It analyzed the frequencies of responses to the 2007 questionnaire compared with those found in that of 2001 through percent variation (PV).
RESULTS: 504 (79.2%) of the services responded to the questionnaire. Almost 100.0% of the respondents reported having essential resources for outpatient care: having at least one doctor, sufficient supplies of antiretroviral drugs, CD4 and viral load tests. Many aspects displayed improvement in 2007 compared to 2001: registry of missed medical appointments (from 18.3 to 27.0%, PV: 47.5%), follow-up appointment within 15 days of starting antiretroviral treatment (from 55.3 to 66.2%, PV: 19.7%) and user's organized participation (from 5.9 to 16.7%, PV: 183.1%). However, some difficulties remained: little change in the availability of specialized exams, such as endoscopy, within 15 days, (31.9 to 34.5%, PV: 8.1%) and decreases in indicators such as optimal time access to specialized appointments (55.9 to 34.5% in cardiology, negative PV: 38.3%). Mean time spent in follow-up medical appointments remained low: about 15 minutes (52.5 to 49.5%, negative PV: 5.8%).
CONCLUSIONS: The 2007 assessment revealed that services have essential resources for ambulatory assistance. There was some improvement in many aspects compared to 2001, although some challenges still remain. Little time dedicated to medical appointments may be linked to insufficient number of doctors and/or due to reduced capacity of listening and dialogue. Impaired access to specialized appointments reveals the difficulty local Brazilian Unified Health System facilities have regarding infrastructure.
Descriptors: Acquired Immunodeficiency Syndrome, therapy. HIV Infections. Ambulatory Care. Brazilian Unified Health System. Health Services Research.
OBJETIVO: Evaluar los servicios del Sistema Único de Salud brasileño de asistencia ambulatoria a adultos viviendo con sida en 2007 y comparar con la evaluación de 2001.
MÉTODOS: Los 636 servicios catastrados en el Ministerio de la Salud en 2007 fueron invitados a responder un cuestionario previamente validado (Cuestionario Qualiaids) con 107 preguntas de selección múltiple sobre la organización de la asistencia prestada. Se analizaron las frecuencias de las respuestas de 2007 comparándolas con las obtenidas en 2001 en la forma de variación porcentual (VP).
RESULTADOS: Respondieron el cuestionario 504 (79,2%) servicios. Cerca de 100,0% de los encuestados relataron tener al menos un médico, suministro sin fallas de antirretrovirales y de exámenes CD4 y carga viral. Varios aspectos mostraron mejor desempeño en 2007 al compararse con 2001: registro de número de faltas a la consulta médica (de 18,3 a 27,0%, VP: 47,5%), conseguir consulta en menos de 15 días en el inicio de la terapia antirretroviral (de 55,3 a 66,2%, VP: 19,7%) y participación organizada del usuario (de 5,9 a 16,7%, VP: 183,1%). Se mantuvieron algunas dificultades: pequeña variación en la disponibilidad de exámenes especializados en hasta 15 días, como endoscopia (31,9 a 34,5%, VP: 8,1%) y empeoraron indicadores como tiempo ideal de acceso a consultas especializadas (55,9 a 34,5% en cardiología, VP negativa de 38,3%). El tiempo promedio empleado en las consultas médicas de seguimiento se mantuvo bajo: 15 minutos o menos (52,5 a 49,5%, VP negativa de 5,8%).
CONCLUSIONES: La evaluación de 2007 mostró que los servicios cuentan con los recursos esenciales para la asistencia ambulatoria. Hubo mejorías en muchos aspectos con relación a 2001, pero persisten desafíos. Poco tiempo dedicado a la consulta médica puede estar vinculado al número insuficiente de médicos y/o a la baja capacidad de atención y diálogo. El acceso perjudicado a consultas especializadas muestra la dificultad de las infraestructuras locales del Sistema Único de Salud.
Descriptores: Síndrome de inmunodeficiencia adquirida, terapia. infecciones por vih. atención ambulatorial. sistema único de salud. investigación sobre servicios de salud.
Outpatient care for those living with HIV in Brazil takes place in different health care sites of the SUS (Brazilian Unified Health System): primary care units, specialized outpatient clinics, hospital outpatient services and sites dedicated to sexually transmitted diseases (STD) and AIDS. The Ministry of Health, through the Department of STD/AIDS and Viral Hepatitis produces general guidelines for the services and is responsible for providing antiretroviral drugs and specific laboratory tests (viral load, CD4, genotyping). Apart from these resources, the implementation and organization of services depends on regional and local characteristics of the SUS.
Local-regional distribution and organizational characteristics are diverse, although all are subject to the same set of general directives. Between 1996, when universal access to highly active antiretroviral therapy (HAART) was established, and 2007, the number of sites delivering special care to HIV patients increased from 33 to 636, according to outpatient sites registered by the Ministry of Health's Department of STD/AIDS and Viral Hepatitis.
Between 2001 and 2002, sites in seven Brazilian states were evaluated, based on evaluative research that developed and validated the Qualiaids Questionnaire. The instrument was completed by local managers. Questions included availability of resources, organization of the care process and management.7
The purpose of the questionnaire in supporting local management and defining policies encouraging quality, along with its potential as an instrument for monitoring and evaluating services, motivated the interest of the Ministry of Health in using it with all the sites in the country. The original questionnaire was adapted to be used electronically in 2005, accompanied by an online good practice recommendations guide referring to the dimensions of care assessed in the questionnaire.ª
From 2007, the Qualiaids Questionnaire was adopted as the official quality evaluation instrument of SUS outpatient HIV services. The Ministry of Health conducted the evaluation requesting all sites in the country to answer the questionnaire. The survey report and database, developed in partnership with the Ministry of Health, were forwarded to the state level managers of the STD/AIDS program and disseminated in workshops conducted by national level managers.b This article aimed to assess outpatient care services for adults living with HIV in 2007 and compare it with the 2001 results.
The Qualiaids Questionnaire was validated in 27 sites with different care organizational characteristics and the results of its first application in 2001 were analyzed and disseminated among site managers, decision makers and specialized publications in the field.7,9,10 For the 2007 evaluation, the electronic questionnaire was made available to the sites.
The universe of sites which participated in the 2007 evaluation was defined by a list of 636 outpatient care services provided by the 27 Brazilian federal STD/AIDS coordinators. The sites were invited to complete the questionnaire online, consisting of 107 multiple choice questions: nine questions concerning institutional characteristics, location and site size; 24 on availability of resources; 42 on organization of care and 32 on local care management.
Descriptive analysis was carried out for 464 sites evaluated in 2007 and comparative analysis was carried out for 204 of these sites which had also been assessed in 2001. Percentages of sites were calculated according to their responses regarding three evaluative dimensions: Resources, Care and Management. Sites missing out ten or more questions were excluded from the analysis. Geographical location and site size were not considered in this analysis. No statistical estimates were included in the comparison of percentages obtained in 2001 and in 2007 as they referred to all respondents sites. Percentage variation between the two applications was calculated (P2007 (% 2007)- P2001 (% 2001)/ P2001 *100).
Being able to schedule tests and appointments with specialists in under 15 days was deemed "promptness" This article used administrative data from the database produced by the application of Qualiaids carried out by the Ministry of Health Department of STD/AIDS and Viral Hepatitis, given to the authors of this article as part of a technical cooperation agreement. The health care sites' managers voluntarily agreed to participate.
Of the 636 sites existent in 2007, 504 (79.2%) answered the questionnaire, of which 464 (73.0%) answered at least 97 of the 107 questions in the instrument and constituted the universe analyzed (Table 1).
Of the 464 sites analyzed, 67.2% were located in municipalities with fewer than 400,000 inhabitants.c
Of the sites, 26.7% treated more than 500 patients on antiretroviral therapy (HAART); 36.0% between 101 and 500, 11.6% between 51 and 100 and 24.1% with up to 50 patients. It was in the south and southeast regions where 80.2% of the sites were located, of which 68.6% units were non-exclusive to STD and AIDS care. (Table 1).
There were 91.4% of sites easily accessible by public transport; 82.8% were open to the public five days or more per week, and 78.1% were open eight hours a day or more.
All sites reported the presence of at least one infectious diseases specialist or general physician responsible for HIV/AIDS care; 44.2% had one physician; 46.1% reported that the doctor(s) had five or more years' experience treating HIV/AIDS.
There was at least one infectious diseases specialist in 68.5% sites and the availability of other professionals in the health care team was high, with the exception of dentists (Table 2).
Promptness in scheduling appointments was high for gynecologists (75.0%). Waiting time for other specialist appointments, taking > 45 days or unavailability, was 24.5% for cardiology and 31.9% for psychiatry.
Approximately 75.7% of sites provided, on average, three or more CD4 count tests per patient/year and 71.8% Viral Load. The promptness of the test results was lower: 28.5% and 19.2% respectively.
In 47.2% of the sites the results of HIV diagnosis using Anti-HIV (Elisa) were available promptly, and this was the case in 20.3% care units for the Western-Blot test. Laboratory tests used to monitor the toxicity of the medications requiring more promptness (blood count, blood urea nitrogen, transaminases) were available in 74.8% of the sites. Simple imaging tests, such as x-ray (chest, sinuses, abdomen), had high availability (76.7%), while the more complex ones (such as endoscopy and CT) were less available.
There was high availability for all of the antiretroviral drugs. There were one week gaps in supply for 6.9% of care units for zidovudine+lamivudine, and 0.9% for amprenavir. Trimethoprim-sulfamethoxazole, the most widely used drug for prophylaxis and treating opportunistic infections, was reported unavailable in 3.2% of the sites; 7.5% reported gaps in supply of more than 15 days.
First time patients were seen by a university-level professional on the same day in 72.0% of the sites.
Appointments were scheduled at the beginning of the period for all patients in 45.7%; 14.7% of the sites reported booking appointments in groups of patients and 27.4% made appointments for a specific time for each patient. Up to ten patients were scheduled for four hours/doctor in 65.7% of the sites; 14.4% scheduled 16 or more appointments per doctor for a 4 hours period. The average time for a follow-up consultation was 15 minutes in 49.4% of the sites.
In the event of patients without appointments, 59.3% of the sites would see the patient on the same day (regardless of whether there was space in the diary); 35.1% of the care units kept "free appointments" for this kind of eventuality.
Clinical complications are the main reason behind demand for unscheduled appointments in 77.8% of sites, followed by requests for testimonials for the purposes of claiming social benefits (43.1%).
The use of some kind of care guideline form or appointment protocol was reported by 51.5% of the sites for first appointments and 56.3% for tuberculosis/HIV co-infection, pregnant women or occupational accidents.
Of the sites analyzed, 80.2% reported that the patients' adherence to treatment was verified in individual appointments through questioning them about correct medication use. Adherence groups were conducted by 28.9% of sites, 21.3% stated they applied detailed records on medication use and 34.3% used pill counts. In 63.8% of sites the first appointment was scheduled within 15 days of starting HAART.
Asked about usual approach when faced with patients expressing the desire to have children, 80.6% of the sites reported providing guidance on the risks and discussing the best time for conception. Routine referral to a gynecologist (even in the absence of complaints) was related by 33.4%.
Approximately 91.2% of the sites had a technical coordinator and in 45.7% of the sites the technical coordination was a fulltime task; 33.0% of managers had been working in management for more than six years and 53.0% had management training.
The most commonly registered data were: number of patients using antiretroviral drugs (92.7%), number of appointments/procedures (89.0%) and appointments per doctor (88.1%). Lower frequencies were observed for systematically recording number of missed appointments per patient (27.0%) and recording the first appointment of the year (38.7%).
Technical or administrative supervision was reported in 56.2% of the sites, and regular staff meetings in 36.0%; 34.1% reported some type of activity aimed at minimizing professionals' suffering.
Meetings in which patients participated were reported by 16.4% of the sites; 40.5% stated that there was some kind of organized patient participation in finding solutions to problems.
Among the most commonly mentioned managerial difficulties were: specialist referrals (58.0%), hiring university level (56.0%) and medium level (42.0%) human resources, finding referral beds for hospitalizations (42.0%), acquiring consumable (36.6%) and permanent (36.9%) resources and the supply of drugs against opportunistic infections (32.1%).
The responses of the 204 sites which completed the Qualiaids Questionnaire in 2001 and 2007 were compared.
Three or more viral load tests per patient/year were more widely available, with a percentage variation (PV) of 90.5%. On the other hand, promptness of referrals to all of the specialists investigated worsened, with PV between 38% and 40% for psychiatry, neurology, cardiology and general surgery (Table 3).
Indicators of care organization showed an increase in the number of sites which booked appointments setting specific time (PV of 29.2%); in the percentage of sites which reported conducting adherence groups (PV 65.7%) and in the percentage of those which scheduled appointments within 15 days of starting antiretroviral therapy (PV of 59.0%). Routine gynecology appointments (regardless of complaints) remained practically unchanged, with a percentage of 37.3% in 2001 and 36.3% in 2007 (Table 4).
Clinical complications, the need to apply for social benefits, running out of medication and missing the previous appointment were the main reasons patients sought unscheduled appointments in 2001. In 2007, all of these percentages were lower, with negative PV between 19.5% and 33.0%. Clinical complications and applying for social benefits remained the first and second most common reasons, respectively (Table 4).
The referral flow remained centralized around the doctors. There was a 15.7% variation in the number of sites whose professionals carried out the referrals for the entire team (Table 4).
The proportion of meetings to discuss cases remained low, with a small, positive variation. Regular staff meetings maintained similar proportions between 2001 and 2007 (Table 5).
A marked increase in patient participation was observed in 110 sites in 2007, especially through using a management council or similar (PV of 183.1%).
There was a significant decrease in the proportion of sites that reported challenges in accessing viral load and CD4 count tests (negative PV of 80.4%). Difficulties in hospital referrals and with supplies of opportunistic infections medication have also decreased.
Referral to specialists and hiring higher level staff remained the most commonly reported difficulties, with little variation between the two assessments.
The 2007 evaluation demonstrated that the care unit network had the resources essential to outpatient care: doctors, antiretroviral drugs supplies, availability of the most important tests for monitoring cases. The majority also reported having a multidisciplinary teams and access to simple imaging tests.
There are, however, significant shortcomings, such as the low availability of dentists, present in just over half the sites.
Viral load and CD4 tests are widely available and provided by the Federal Government, in contrast to other kinds of tests, which depend on the local public health network (state or municipality responsibilities). Access to specialists also depends on regional and local networks, and is difficult or non-existent in many sites.
Although the majority of sites are open all day long, functional access is undermined in those units that schedule all patients (or group of patients) at the beginning of the shift. This frequently means long waiting times, typically associated with lower levels of patient satisfaction.
It is also worrying that follow-up appointments are expected to take less than 15 minutes in many sites. A similar study carried out with 21 American sites showed an average appointment time of 20 minutes, varying between 15 and 40, for this type of appointment.16
It is probable that, for some sites, the main problem is a lack of doctors, given the shortages and poor distribution of this professional, especially in public health services. Moreover, "squeezing in" patients who do not have appointments may contribute to doctors being over worked: 35.1% of sites reported having some kind of system (such as keeping slots free), although 59.3% stated that all patients without an appointment were seen.
Even with an appropriate number of hours/doctor, some sites may have professionals who work fewer hours than those officially contracted. This is a problem frequently experienced by health care managers, although no empirical studies on this topic exist. It is possible that, faced with the impossibility of changing this situation, some managers reach an informal agreement with doctors on the maximum number of hours within their actual time spent at the site. Thus, they guarantee to meet care demands, although the length of appointments are reduced and may compromise the quality of care. Doctors with shorter shifts may be the reason for scheduling all patients or group of patients at the beginning of the shift.
There is no available evidence to support guidelines for length of consultation on HIV/AIDS. In general clinical practice, the time varies between countries and site types. It is, however, considered at least a proxy measure of quality.2,14 Short consultations suggest insufficient technical quality and little capacity for listening and dialogue. This is a similar profile to what Campos describes as "debased practice",1 and typically observed in unscheduled appointments.
Dealing with unscheduled cases, far from being a fast and "simple" health care alternative, is a technologically complex and prioritized moment, both from the point of view of diagnosing significant clinical complications and dealing with patients who have missed appointments, a group at greater risk of non-adherence.d
Even seen as a proxy indicator of quality, studies emphasize the importance of the professionals' experience and specialization in obtaining good clinical outcomes,4,5,13 something which has been highlighted since the beginning of the epidemic.3 Our study observed that almost half of the doctors has five or more years' experience, which could be considered as adequate experience of dealing with HIV patients. On the other hand, around a third of sites did not have an infectious disease specialist, whose original training in HIV/AIDS is more specialized.
Qualified and flexible support systems (mentoring) contribute to the improvement of the quality of medical care in de-centralized systems.6 However, establishing such systems appropriately is not observed in most outpatient units of the SUS. In this study, the majority of HIV/AIDS sites reported that they had no systematic way of supervising or supporting professionals.
The opportunity to exchange opinions and discuss cases in day-to-day professional interaction, a traditional way of learning, is not viable in almost 40% of sites, in which there was one doctor responsible for monitoring HIV patients. Sites with only one doctor are associated with lower quality of care.11
As important as the qualification of the professionals who care for the patients, effective technical - and not just bureaucratic - management enhances the effectiveness of health care: establishing and monitoring care routines, assessing and monitoring work, integrating the team and enhancing communication with patients and third parties.12
It is the responsibility of management to ensure minimum structures of care for all patients and, simultaneously, for those who are more vulnerable to falling ill. Among those are the groups most at risk of non-adherence to treatment. It is positive that most of the sites routinely schedule appointments at shorter intervals at the beginning or HAART, a period crucial to adherence.8,e
A large number of sites show negative characteristics for typical management activities. In terms of assessment and monitoring, a high percentage of sites do not calculate traditional performance indicators, such as coverage and concentration of activities, due to a lack of recorded data. In terms of team integration, there are insufficient formal internal communication mechanisms and a lack of support and supervision activities for the professionals. Regarding social control and participation, there are not enough mechanisms for patient participation and communication with civil society organizations.
Comparison of the 204 sites, which completed both evaluations, showed improvements but also significant persistent deficiencies.
The improvement in the availability of resources under federal management stands out among the characteristics evaluated: there was a significant improvement in the availability of CD4 count and viral load tests. Antiretroviral supply became more timely: running out of medication is no longer a factor for patients seeking unscheduled appointments.
The comparison showed some negative characteristics that persisted in managing health care, at a local level. Routine referral to a gynecologist is still not adequate, in spite of the importance of the women's health issue in the program guidelines.f Multi-professional team integration is flawed, indicated by the large proportions of doctor-centered referrals and the lack of regular staff meetings. Moreover, there are few sites which seek to improve accessibility by scheduling appointments at a specific time.
The consolidation of broader access to essential resources, such as new medication and tests, reiterates the Brazilian AIDS program's justified reputation for good performance, compared to other middle income countries at that time.15
HIV patient care in Brazil brought up ethical commitments and technological propositions that serve as examples to the Brazilian Health System. It is, however, in health care professionals' and local managers' day-to-day work that these proposals are realized. Standardizing, encouraging and monitoring the quality of this work is the challenge faced by these surveyed sites, and the health care system as a whole.
This study has some limitations: it does not include indicators of results and is based on local managers' responses. The sample was determined by voluntary participation and 27% of the sites did not answer the questionnaire or did not fulfill all of the requisites, and could not be included in the analysis. The only known characteristics of the non-respondents were the states and municipalities in which they were located. These aspects are not sufficient to understand possible differences in relation to the universe of the study, given the heterogeneity of sites within the same geographical area. On the other hand, the sites analyzed include the diversity of locations and types of delivery of care to HIV patients in Brazil.
As it examines process indicators, this assessment values the necessary conditions for the long-term care of a chronic condition, such as AIDS currently is. Assessing and monitoring essential characteristics of structure and process is the first step in a broader initiative to improve health care quality. The possibility, albeit limited, of assessing the evolution of these characteristics over a six-year-period is unprecedented in SUS outpatient care.
The monitoring and evaluation initiative led by the Ministry of Health Department of STD/AIDS and Viral Hepatitis involved the participation of states and municipalities (capitals and larger cities) AIDS program coordinators in two rounds of analysis. This process was repeated in some states and municipalities with local health teams. The analysis resulting from this evaluation promoted conditions for changes in the management and organization process, thus completing the "assessment cycle".
1. Campos GWS. Saúde Paidéia. São Paulo: Hucitec; 2003.
2. Deveugele M, van den Derese A, Brink-Muinen AVD, Bensing J, De Maeseneer J. Consultation length in general practice: cross sectional study in six European countries. BMJ. 2002;325(7326):472. DOI:10.1136/bmj.325.7362.472
3. Kitahara MM, Koepsell TD, Deyo RA, Maxwell CL, Dodge WT, Wagner EH. Physicians' experience with the acquired immunodeficiency syndrome as a factor in patients' survival. N Engl J Med. 1996;334(11):701-6. DOI:10.1056/NEJM199603143341106
4. Landon BE, Wilson IB, Cohn SE, Fichtenbaum CJ, Wong MD, WengerNS, et al. Physician specialization and antiretroviral therapy for HIV. J Gen Intern Med. 2003;18(4):233-41.
5. Landon BE, Wilson IB, McInnes K, Landrum MB, Hirschhorn LR, Marsden PV, et al. Physician specialization and the quality of care for human immunodeficiency virus infection. Arch Intern Med. 2005;165(10):1133-9. DOI:10.1001/archinte.165.10.1133
6. Llibre JM, Domingo P, Pozo MA, Miralles C, Galindo MJ, Viciana I, et al. Long-distance interactive expert advice in highly treatment-experienced HIV-infected patients. J AntimicrobChemother. 2008;61(1):206-9. DOI:10.1093/jac/dkm438
7. Melchior R, Nemes MIB, Battistella MI, Basso CR, Castanheira ERL, Alves MTSSB,et al. Avaliação da estrutura organizacional da assistência ambulatorial em HIV/Aids no Brasil. RevSaude Publica. 2006;40(1):143-51. DOI:10.1590/S0034-89102006000100022
8. Melchior R, Nemes MIB, Alencar TMD, Buchalla CM. Desafios da adesão ao tratamento de pessoas vivendo com HIV/Aids no Brasil. RevSaude Publica. 2007;41(Supl 2):87-93. DOI:10.1590/S0034-89102007000900014
9. Nemes MIB, Castanheira ERL, Melchior R, Alves MTSSB, Basso CR. Avaliação da qualidade da assistência no programa de AIDS: questões para a investigação em serviços de saúde no Brasil. Cad SaudePublica. 2004;20(Supl 2):S310- 21. DOI:10.1590/S0102-311X2004000800024
10. Nemes MIB, Melchior R, Basso CR, Castanheira ERL, Alves MTSSB, Conway S. The variability and predictors of quality of AIDS care services in Brazil. BMC Health Serv Res. 2009;9:51. DOI:10.1186/1472-6963-9-51
11. Rodriguez HP, Marsden PV, Landon BE, Wilson IB, Cleary PD. The effect of care team composition on the quality of HIV care. Med Care Res Rev. 2008;65(1):88-113. DOI:10.1177/1077558707310258
12. Schraiber LB, Peduzzi M, Sala A, Nemes MIB, Castanheira ERL, Kon R. Planejamento, gestão e avaliação em saúde: identificando problemas. CiencSaudeColetiva. 1999;4(2):221-42. DOI:10.1590/S1413-81231999000200002
13. Valenti WM. The HIV specialist improves quality of care and outcomes. AIDS Read. 2002;12(5):202-5.
14. Wilson A, Childs S. The relationship between consultation length, process and outcomes in general practice: a systematic review. Br J Gen Pract. 2002;52(485):1012-20.
15. World Health Organization. Report on the global HIV/AIDS epidemic 2008: executive summary. Geneva; 2008.
16. Yehia BR, Gebo KA, Hicks PB, Korthuis T, Moore RD, Ridore M, et al. Structures of care in the clinics of the HIV Research Network (HIVRN). AIDS Patient Care STDS. 2008;22(12):1007-13. DOI:10.1089/apc.2008.0093