Nurse-based rapid assessment clinics in Western Kenya may improve survival and clinical retention of very sick patients beginning combination antiretroviral therapy (ART), Paula Braitstein and colleagues reported at the HIV Implementers' Meeting in Namibia last month.
Launched in 2001 the USAID-Academic Model Providing Access to Healthcare (AMPATH) has provided services to over 90,000 men, women and children with HIV in 18 parent and 10 satellite clinics in urban and rural areas throughout Western Kenya.
High Risk Express Care began as a pilot project in March 2007 in four high-volume clinics. Reducing mortality and loss to follow up in HIV-infected adults with CD4 counts below 100 cells/mm³ when beginning ART, as well as increasing clinic capacity without additional costs, were the primary goals.
By June 2008 the project had been rolled out to 18 clinics.
Routine care for patients beginning ART involves a clinical officer seeing the patient at every visit and prescribing ART. Monthly visits are scheduled unless clinical indications determine otherwise.
High Risk Express Care for patients beginning cART involves a clinical officer seeing the patient and prescribing ART. The patient is then referred to Express Care (EC) upon ART initiation. The clinical officer will see the patient on a monthly basis. In the interim weeks over a period of three months a nurse will either see the patient in the clinic or talk to them over the phone. Vital signs are taken and a rapid symptom assessment is done each time, with immediate referral to a clinical officer if symptoms call for it.
To assess the comparative impact of High Risk Express Care and routine care on clinical outcomes a retrospective observational study was undertaken. Criteria for inclusion included: beginning ART, having a CD4 count below 100 cells/mm³ and being 14 years of age and over. Endpoints were mortality and loss to follow up defined as absent from the clinic for at least 3 months without evidence of patient death.
Over a period of 10.5 months (March 1, 2007 and January 15, 2008) 2601 patients with a CD4 count below 100 began cART. A total of 14 out of the 28 clinics had begun HREC with a corresponding total of 378 (14.5%) eligible patients enrolled. Median cell count at initiation for the routine care group was 44, compared with 47 in the HREC group. The probability of remaining alive after 300 days was 95% for those in express Care and 91% for those in routine care. The probability of remaining alive and in care after 300 days was 86% for those in express care and 75% for those in routine care. In both cases the results were statistically significant.
Concern over selection bias in terms of clinics and patients selected for express care, use of cotrimoxazole (Septrin) and provider bias regarding adherence to protocols prompted the researchers to perform a sub-analysis which was restricted to clinics which had initiated express care. Eligibility criteria for patients in this sub-analysis included: initiation of ART after express care was initiated in the clinic and being eligible for express care (a CD4 count of ≤100 cells/mm³).
A total of 715 patients were included, 336 (46.9%) in express care and 379 (53.1%) in routine care. A lower proportion was on TB treatment and a higher proportion attended urban clinics than in the initial analysis. 98% (EC) versus 91% (RC) were using cotrimoxazole at initiation of cART.
Ninety-six per cent of patients in express care and 89/90% in RC were alive after 300 days; 85% in EC and 76% in RC were alive and in care after 300 days. Adjustment for all factors (gender, age, CD4 at cArt initiation, treatment for tuberculosis at cART initiation, clinic, use of cotrimoxazole or dapsone at ART initiation, WHO clinical stage at ART initiation, and time taken to get to clinic) showed a 60% decrease in mortality for those in EC. Adjustment for the same factors indicated that those in express care were half as likely to become lost to follow up (AHR 0.45, 95% CI:0.27-0.77).
Those in express care were seen by a dedicated nurse team at the clinic, helping ease clinic congestion. No cost-effectiveness analysis was undertaken.
The authors note that study limitations include the fact that the data are observational, and determination of outcomes is incomplete. They also note that just because clinical protocols exist, they may not be fully implemented, which could lead to an underestimate of the impact of express care when fully implemented.. Adherence to other protocols, for example cotrimoxazole prophylaxis, may also affect outcomes.
High risk express care appears to improve clinical retention and reduces mortality, but there is uncertainty as to whether outcomes are due to early identification, improved adherence or the dedication of nurses.
Paula Braitstein of AMPATH said that paying close attention and dealing with issues rapidly contributed to the outcomes. Symptoms were identified within days, adherence barriers readily identified, and referral took place immediately when problems arose. No shows are followed-up, and there is an active outreach programme led by HIV-positive people. In this specific setting nurses were underutilised, she noted, so their time so could be dedicated to express care, and consequently their workload was not affected, but this is not the norm in Kenya. She concluded that this model of care is “possibly generalisable and intuitively makes sense”.
Reference
Braitstein, P et al. High Risk Express Care: a novel care model to reduce early mortality among high risk HIV-infected patients initiating combination antiretroviral treatment. HIV Implementers’ Meeting, Namibia, abstract 1556, June 2009.