A clinical audit of provider-initiated HIV counselling and testing in a gynaecological ward of a district hospital in Kwazulu-Natal, South Africa.
A recent systematic review and a number of observational studies [5-7] report that PICT in low- and middle-income countries (LMICs) has increased HIV testing rates. Most of these studies on increased HIV testing have been carried out in antenatal clinics, tubercolosis centres and sexually transmitted disease clinics. There are few data on whether PICT has been integrated into routine care provided by health facilities catering for the general population.
Our impression is that PICT is patchy at best in a gynaecological ward in a district hospital in KwaZulu-Natal, South Africa. We therefore carried out an audit of PICT among women admitted to the ward.
A retrospective chart review was performed of all patients admitted to the gynaecological ward who stayed for a period of [less than or equal to]3 days over an 8-month period between 1 October 2011 and 31 May 2012. The charts of individual patients were reviewed immediately after discharge, and relevant data such as demographic information and whether the patient was offered HIV testing were recorded in a structured format. Institutional ethics (BREC 196/12) and hospital regulatory permission was obtained. The data were analysed using simple descriptive statistics.
The general policy at the study site is to offer PICT to all women admitted to the gynaecology ward. In this model of HIV testing, all healthcare workers are encouraged to initiate the counselling and testing process by suggesting the need for a test. The patient can opt out if she desires. Detailed counselling is provided once the HIV result is known, and referral to a dedicated HIV clinic is arranged. Patients who have been tested previously and state that they had a negative result are offered re-testing. The rapid HIV test used is the First Response (Premier Medical Corp Ltd, India). If the patient tests positive, a second HIV rapid test, the Determine (Alere Medical, Co. Ltd, Japan), is used as a confirmatory test. If both rapid tests are positive, the result is regarded as confirmation of the presence of HIV infection. If the second test is negative, a laboratory enzyme-linked immunosorbent assay is performed to confirm the result. All healthcare professionals were expected to document whether testing was offered, and if it was, whether the patient agreed to testing. The result of the rapid HIV test was also documented in the patient's hospital records. Healthcare professionals were also told to record whether a patient declined testing. The nursing staff kept a separate register of all patients who had an HIV test. [CD4.sup.+] counts were not obtained immediately, but patients were referred to a HIV clinic with appropriate information. At the time of the study, the [CD4.sup.+] count evaluation took 3-5 days to be processed, and some patients were discharged from hospital before their results became available.
All HIV-positive patients were clinically staged, and those who were clinically stage III and IV (World Health Organization staging) had further counselling, the relevant HIV investigations, appropriate referral and clinical management. All HIV-positive patients irrespective of clinical stage were referred to dedicated HIV clinics.
The district hospital is in an urban setting in the Durban metropolitan area and is mainly attended by lower socio-economic population groups. The overall HIV prevalence among antenatal women aged 15 - 49 years in KwaZulu-Natal is estimated at 39.5%.  Patients who stayed in the ward for >3 days were excluded from the audit because we believed that the longer the duration of hospital stay, the greater the chances were of PICT being initiated.
Over the 8-month study period, 1 019 patients were admitted to the ward and stayed for [less than or equal to]3 days; 1 014 case files were evaluated (5 files were missing). The mean age of all patients ([+ or -] standard deviation (SD)) was 23.4 ([+ or -] 6.3) years (range 18-54). The majority (65%) were in the 18-29-year age group. The mean parity was 1.08 ([+ or -] 1.05) (range 0-4) and the median 2. The mean hospital stay was 2.35 ([+ or -] 0.48) days (range 2-3).
HIV counselling and testing
Of the 1 014 case files audited, 451 (44.4%) had documented evidence of patients reporting that they had been tested previously. Fig. 1 shows that 916 patients should have been offered PICT (353 who had previously tested negative and 563 who had not had HIV testing). Of these, 157 (17.1%) were offered PITC and 116 (73.9% of those who were offered) accepted; 45 of these (38.8%) tested positive (Fig. 1).
Table 1 shows details of HIV testing in relation to the various clinical diagnoses recorded in the files. Miscarriages were the commonest indication for admission. This group comprised 599 (59.1%) of all admissions, and HIV testing was offered to 107 (17.9%) of these patients. Thirty-three patients with Bartholin's abscesses should have been offered testing, but only one was offered it (Table 1).
Of 1 014 gynaecological patients with acute admissions of [less than or equal to]3 days, 451 had been tested for HIV previously and 98 (21.7%) were HIV-positive (Fig. 1). This prevalence of HIV positivity is lower than in the National HIV Antenatal Prevalence Survey,  where the average figure for pregnant women is approximately 29%. A total of 916 women (90.3%) had not had an HIV test performed.
In our audit, only 157 (17.1%) of these 916 patients, who should have been offered HIV testing, were offered it. This confirms our observation that PICT is not being properly implemented in settings in which women present with acute gynaecological disorders and there is a rapid turnover of patients due to the shortage of hospital beds and possibly lack of trained staff and counsellors. There is also high staff turnover, so ongoing training on standard clinical guidelines may be lacking.
The above operational issues concerning PICT should be taken into account if we are to test more individuals for HIV and start treating those who are positive as early as possible. One reason why so many patients were not offered testing may be that the date of the previous HIV test was not established when patients stated that they had been tested. These could be regarded as missed opportunities for retesting, especially for high-risk patients such as those with a diagnosis of pelvic inflammatory disease, Bartholin's abscess and cervical cancer. More importantly, it would seem that healthcare workers do not understand the PICT model. In this model, doctors or nurses can suggest to the patient that he/she has a test, and instead of a detailed consent form being worked through and signed, a note in the file indicating that that HIV testing has been suggested and agreed to is sufficient. The healthcare worker then performs the test and the detailed counselling is arranged. The counsellor can then perform this time-consuming but important part of the PICT process. It appears that healthcare workers are not fully informed about this procedure, and are distancing themselves from this responsibility and placing too much emphasis on the counsellors. Integration of the process into routine healthcare and disseminating information to all healthcare workers is essential if we are to increase the numbers of people who have HIV screening and testing.
The study subjects spent an average of 2 days in hospital, and most of them were young and had uncomplicated miscarriages. Only 20% of the patients with incomplete miscarriages and 12.8% of those with ectopic pregnancies were offered testing (Table 1). It is our impression that in an urban setting, patients with a diagnosis of miscarriage ask to be discharged from hospital as soon as possible. Delays in counselling and testing before discharge from hospital may have been a reason why so few were offered PICT, although it is possible that PICT was offered but not recorded in the hospital notes. This is one of the limitations of this retrospective audit. We only included women who stayed in hospital for [less than or equal to]3 days (an arbitrary figure) because we wanted to overcome any biases resulting from the fact that patients who stay in hospital for longer periods may have a greater chance of being offered HIV testing.
Other operational issues such as privacy and confidentiality may have played a role in healthcare workers not offering PICT. Obermeyer and Osborn  reported that nurses noted a lack of private space as a major constraint to discussions before and after HIV testing. Reports have also indicated the lack of private rooms for maintaining privacy and confidentiality as probable obstacles to PICT. [10,11] These resources were certainly lacking at the study site.
Wanyenze et al.,  in a study performed in Uganda, reported that before implementation of PICT only 20% of the patients discharged from a medical ward had received HIV testing, but after the initiation of PICT 98% agreed to a test and 81% were tested for the first time. The HIV prevalence was 25%. We did not have baseline data, but before the recommendation of PICT, an audit of HIV testing among women seeking termination of pregnancy at our study site showed low rates of HIV testing.  Monitoring and evaluation of PICT in general wards, continuing training of all healthcare workers and counsellors, provision of staff and sufficient privacy are essential if we are to increase the number of people tested at the point of healthcare contact.
In our study, 73.9% of those who were offered PICT accepted testing. This is a reasonably high figure in a setting of acute gynaecological emergencies. More importantly, 38.8% were HIV-infected. Although this is a 'select' group that would be expected to have a higher prevalence than that in the general population (17.8%), the figure emphasises the need for PICT at all clinical contacts in public sector health facilities. 
Defaulting from return visits is common in LMICs. Painter et al.  reported that 72% of their patients accepted HIV testing, but only 45% returned to collect their results. Rapid HIV testing in the ward would overcome this problem. It could be argued that additional staff would probably be required for this, but we must remember that doctors and nurses have the knowledge and skills to assist in the initial counselling and testing. Much of the in-depth, time-consuming counselling can be done by the trained counsellors. These operational issues require urgent attention.
Although, as stated above, the 73.9% response rate to PICT in our study suggests a high level of acceptability, other studies have revealed that patients may feel forced into HIV testing, or feel compelled to agree to it. [15,16] Our study did not take these factors into account, and further investigations in general wards with rapid patient turnover are required. Patients may feel compelled to agree to testing if they want to be discharged from hospital as soon as possible.
If the operational issues mentioned above are attended to, widespread implementation of PICT may be able to deliver the large-scale increase in HIV testing and counselling that is required to initiate early treatment and decrease HIV transmission at a population level.
[1.] World Health Organization/UNAIDS/UNICEF. Towards Universal Access: Scaling Up Priority HIV/AIDS Intervention in the Health Sector: Progress Report 2010. Geneva: WHO, 2010.
[2.] Hammer SM. Antiretroviral treatment as prevention. N Engl J Med 2011;365(6):561-562. [http:// dx.doi.org/10.1056/NEJMe1107487]
[3.] Cohen MS, Chen YQ, McCauley J, et al. Prevention of HIV-1 infection with early antiretroviral treatment. N Engl J Med 2011;365(6):493-505. [http://dx.doi.org/10.1056/NEJMoa1105243]
[4.] Dinh TH, Goga A, Jackson D, et al. Impact of South Africa's PMTCT programs on perinatal HIV transmission: Results of the 1st year implementing the 2010 WHO recommended guidelines. Abstract book, 4th International Workshop on HIV Pediatrics, Washington, DC, USA, 20-21 July 2012.
[5.] Kennedy CE, Fonner VA, Sweat MD, Okero FA, Baggaley R, O'Reilly KR. Provider initiated HIV testing and counselling in low and middle income countries: A systematic review. AIDS Behav 2013;17(5):1571-1590. [http://dx.doi.org/10.1007/s10461-012-0241-y]
[6.] Dalal S, Lee C-W, Farirai T, et al. Provider-initiated HIV testing and counseling: Increased uptake in two public community health centers in South Africa and implications for scale-up. PLoS One 2011;6(11):e27293. [http://dx.doi.org/10.1371/journal.pone.0027293]
[7.] Leon N, Naidoo P, Mathews C, Lewin S, Lombard C. The impact of provider-initiated (opt-out) HIV testing and counselling of patients with sexually transmitted infections in Cape Town, South Africa: A controlled trial. Implement Sci 2010;5:8. [http://dx.doi.org/10.1186/1748-5908-5-8]
[8.] National Department of Health. The 2010 National Antenatal Sentinel HIV and Syphilis Prevalence Survey in South Africa. Pretoria: DOH, 2011.
[9.] Obermeyer C, Osborn M. The utilisation of testing and counselling for HIV: A review of the social and behavioural evidence. Am J Public Health 2007;97(10):1762-1774. [http://dx.doi.org/10.2105/ AJPH.2006.096263]
[10.] Bell E, Mthembu P, O'Sullivan S, Moody K. Sexual and reproductive health services and HIV testing: Perspectives and experiences of women and men living with HIV/AIDS. Reprod Health Matters 2007;15(S29):113-135. [http://dx.doi.org/10.1016/S0968-8080(07)29029-9]
[11.] Greeff M, Phetlhu R, Makoae L, et al. Disclosure of HIV status: Experiences and perceptions of persons living with HIV/AIDS and nurses involved in their care. Qual Health Res 2008;18(3):311-324. [http://dx.doi.org/10.1177/1049732307311118]
[12.] Wanyenze R, Nawavvu C, Namale A, et al. Acceptability of routine HIV counselling and testing, and HIV seroprevalence in Ugandan hospitals. Bull World Health Organ 2008;86(4):302-309. [http://dx.doi.org/10.2471/BLT.07.042580]
[13.] Devjee J, Khedun S, Moodley J. A study of surgical termination of pregnancy at a level 1 health facility in South Africa. S Afr Fam Pract 2012;54(1):72-76.
[14.] Painter T, Diaby K, Matia D, et al. Women's reasons for not participating in follow up visits before starting short course antiretroviral prophylaxis for prevention of mother to child transmission of HIV: A qualitative interview study. BMJ 2004;329:543-546. [http://dx.doi. org/10.1136/bmj.329.7465.543]
[15.] Karim Q, Karim S, Coovadia H, Susser M. Informed consent for HIV testing in a South African hospital: Is it truly informed and truly voluntary? Am J Public Health 1998;88(4):637-640.
[16.] Medley A, Maman S, Kass N. The complexities of the opt out approach to HIV testing within the ante-natal setting: An experience from Uganda. Presented at the 16th International AIDS Conference, Toronto, Ontario, Canada, 13-18 August 2006. Abstract 2199453.
J. Moodley, (1) MB ChB, FCOG, FRCOG (UK), MD; M. Bryan, (2) MB ChB; K. Tunkyi, (2) MB ChB, FCOG; S.M. Khedun, (1) MMedSci
(1) Women's Health and HIV Research Group, Department of Obstetrics and Gynaecology, Nelson R Mandela School of Medicine, University of KwaZulu-Natal, Durban, South Africa
(2) Department of Obstetrics and Gynaecology, Addington Hospital, Durban, South Africa
Corresponding author: J Moodley (email@example.com)
Table 1. PICT in relation to clinical diagnoses Tested for HIV previously Diagnosis N Positive, n Negative, n Miscarriages Incomplete 355 25 77 Inevitable 64 8 16 Complete 84 12 19 Threatened 81 9 16 Septic 15 -- 2 Ectopic pregnancy 86 -- 34 HMB 42 4 26 Wound sepsis 13 2 11 Cancer Cervix 37 6 23 Endometrium 3 0 3 Bartholin's abscess 38 5 7 Fibroid uterus 23 1 13 PID 124 26 62 Others 49 0 44 Total 1 014 98 353 Diagnosis N Not tested Total not for HIV tested for previously, n HIV, n Miscarriages Incomplete 355 253 330 Inevitable 64 40 56 Complete 84 53 72 Threatened 81 56 72 Septic 15 13 15 Ectopic pregnancy 86 52 86 HMB 42 12 38 Wound sepsis 13 -- 11 Cancer Cervix 37 8 31 Endometrium 3 -- 3 Bartholin's abscess 38 26 33 Fibroid uterus 23 9 22 PID 124 36 98 Others 49 5 49 Total 1 014 563 916 PICT Diagnosis N Offered, Accepted, n n Miscarriages Incomplete 355 66 66 Inevitable 64 18 12 Complete 84 3 -- Threatened 81 8 4 Septic 15 12 10 Ectopic pregnancy 86 11 2 HMB 42 5 -- Wound sepsis 13 -- -- Cancer Cervix 37 5 2 Endometrium 3 -- -- Bartholin's abscess 38 1 0 Fibroid uterus 23 1 1 PID 124 26 18 Others 49 1 1 Total 1 014 157 116 PICT Diagnosis N Declined, Not offered, n n (% of total not tested) Miscarriages Incomplete 355 -- 264 (80.0) Inevitable 64 6 38 (67.9) Complete 84 3 69 (95.8) Threatened 81 4 64 (88.9) Septic 15 2 3 (20.0) Ectopic pregnancy 86 9 75 (87.2) HMB 42 5 33 (86.8) Wound sepsis 13 -- 11 (100.0) Cancer Cervix 37 3 26 (83.9) Endometrium 3 -- 3 (100.0) Bartholin's abscess 38 1 32 (97.0) Fibroid uterus 23 -- 21 (95.4) PID 124 8 72 (73.4) Others 49 -- 48 (98.0) Total 1 014 41 759 (82.9) PICT = provider-initiated counselling and testing; HMB = heavy menstrual bleeding; PID = pelvic inflammatory disease.
|Printer friendly Cite/link Email Feedback|
|Author:||Moodley, J.; Bryan, M.; Tunkyi, K.; Khedun, S.M.|
|Publication:||South African Journal of Obstetrics and Gynaecology|
|Date:||Apr 1, 2014|
|Previous Article:||Asymptomatic bacteriuria in women attending an antenatal clinic at a tertiary care centre.|
|Next Article:||Postpartum anal incontinence in a resource-constrained setting: prevalence and obstetric risk factors.|