Practical experiences of integration

The train for integration has already left the station, judging from various presentations made during the 2011 International AIDS Society conference, as well as a large number of other posters on integration presented at the conference.

Areas of integration between HIV services and other health care needs already underway reflect a number of different approaches to integration, and include:

• Integration of malaria prevention and treatment with HIV testing in antenatal care
• Delivery of a basic care package for people with HIV that includes malaria prevention and safe water interventions.
• Delivery of a basic care package for people with HIV that addresses all the co-morbidities identified in the target population
• Leveraging of the HIV programme to improve diabetes care for the general population;

These studies demonstrated that integration is best tailored to local opportunities and needs. These projects are also generally a work in progress — and not every aspect always goes according to plan.

Integrating HIV, family planning and malaria prevention and care

In post-conflict Northern Uganda, there is a high rate of maternal mortality 440/100,000 (2008 UNICEF), and there is a low uptake of many of the services that are critical to improving maternal health.¹ For example, the uptake of family planning services was around 18% in 2006, coverage with two doses of intermittent preventive treatment (IPT2) for malaria stood at only 33%, and the uptake of HIV testing was around 64% (2009 UNICEF).

“These services were generally offered separately and many opportunities to better serve eligible women get missed,” according to Dr Andrew Ocero of JSI and the Northern Uganda Malaria AIDS TB Program, who described a study on the uptake of these services once they were integrated together and provided at antenatal clinics in Northern Uganda.

Two population-based surveys were conducted, one in 2008 (before integration) and the other in 2010, that included women who had delivered in the past two years. The sample size was 855 women in both surveys. The uptake of reproductive health services was examined and compared during the two survey periods.

In 2010, there was an increase in service uptake for each of the critical interventions. The percentage of women who tested for HIV during ANC increased significantly from 71% in 2008 to 88% in 2010; the percentage who received their results went up from 67% to 84%, while the percentage who disclosed their result went up from 62% to 81%.

There was also a significant increase in the uptake of malaria interventions. Ownership of at least one mosquito bed net rose from 56% in 2008 to 91% in 2010; the proportion of those who received at least two doses of intermittent preventive treatment (IPT) for malaria increased from 59% to 70%. Those who tested for HIV were slightly more likely to have received two or more doses of IPT2 (OR 1.62). Also, in 2010, 71% reported having slept under a mosquito bed net throughout their pregnancy, compared with 48% in 2008.

Although still low, 19% of all women interviewed reported using a modern method of family planning compared to 13% in 2008.

“We certainly learnt that there was an increase in service uptake in all the various areas - malaria, family planning, and HIV testing… and PMTCT services were more integrated. However they were challenged by the frequent stock-outs of PMTCT drugs that were being provided by the Ministry of Health,” said Dr Ocero. In addition, “we thought there was still a missed opportunity by women not routinely attending the post-natal clinics.”

Among the policy implications, Dr Ocero concluded that the provision of free insecticide treated bednets or the provision of malarial services at the ANC could act as an incentive to improve the uptake of antenatal services. He also noted that having community-based peer supporters who would help follow up and provided community sensitisation of women for the antenatal services was critical.

Integrating HIV and TB and maternal care

“Intelligent use of data” is needed to select good targets for integration, according to one poster presentation, which suggested that HIV and TB services should be integrated into maternal care in Nigeria because women are disproportionately less likely to be aware of HIV and TB than men.² The conclusion was based on data drawn from a 2008 Demographic and Health Survey collected through a 2-stage stratified random sample of 33,358 Nigerian women aged 15-49 years.

11.68% of Nigerian women had not heard of AIDS with 9 states (24%) accounting for over 50% of the total burden. 28.93% have not heard of tuberculosis (TB) with 11 states (29.7%) accounting for over 50% of the burden. 31.06% have not heard of AIDS and/or TB with 11 states (29.7%) accounting for over 50% of the burden. In all categories analysed, younger age, no education, rural residence, and poverty were found to be significant predictors of low knowledge about HIV.

However, between 29% and 39.7% of these women had accessed ANC care.

“This study shows, in the face of limited resources, significant achievements can be recorded by focusing efforts on ANC/Delivery rather than FP services,” wrote the study’s author Osondu Ogbuoji.

Integrating basic care packages

Basic packages of preventive care (BCPs) can include a variety of interventions that target a variety of infections, chronic illnesses and other health risks.

After several studies in Uganda had previously shown benefits from the delivery of a basic package care and prevention package for people living with HIV, it became a major objective in the Ugandan National Strategic Plan, according to Simon Sensalire, of the Program for Accessible Health Communication and Education in Kampala, Uganda.³

Sensalire gave a presentation on the results of a cross-sectional study, evaluating the effects of integrating the basic care package into care and treatment services for people living with HIV in that country.

The basic care package included two long lasting treated mosquito bed nets (ITBN), a safe water system (WaterGuard solution), condoms and information education and communication (IEC) materials on how to prevent opportunistic infections (OIs) and HIV transmission.

Cotrimoxazole (CTX) was not included in the kit, but was distributed to the same population by the National AIDS Programme.

In addition to the BCP, trained health providers and peer educators provided multiple channels of communication on the prevention of opportunistic infections, family planning, palliative care, TB/HIV and nutrition during routine HIV primary care visits, to people living with HIV.

The cross-sectional survey recruited a nationally representative sample of 2,567 PLHIV at 50 sites in 2010. Findings on the use of the BCP products on the health of PLHIV and other behaviour indicators and how this was affected by the use of communications were compared with a baseline survey conducted in the same target population in 2004/5 (when they would have had to buy some of these components, like the insecticide-treated bed net, for themselves, or boil their water for drinking).

The behavioural indicators include the use of the kit, the consistent use of the condoms among those who are sexually active; sleeping under a treated mosquito bed net every night; treating water with WaterGuard solution; and daily uptake of CTX.

Compared to 2005, the survey found profound improvements in all of these indicators.

Use of the basic care package

Simple provision of the kits improved the behavioural indications, but uptake was significantly increased  when combined with “increased exposure of the information about the BCP kits, specifically the peer education activities,” said Sensalire. Likewise, provision of the kit improved the perceptions among PLHIV that they could always find CTX and condoms, to around 90%. Although perceptions also improved regarding the availability of the insecticide treated bed-net, this was less universal (58. 7%).

“This BCP kit is an essential component for PLHIV and should really be integrated into care and treatment services, especially in poor resource settings [as it] is associated with the uptake of these products for healthy living, but also positive prevention behaviour,” said Sensalire. “The involvement of peer educators, who are PLHIV themselves, backed up by IEC on the BCP kit was associated with improved uptake of these products, at the same time positive prevention behaviour. So this calls for the greater involvement of people living with HIV, in HIV programmes. Finally, the BCP is a valuable approach for promoting public private partnerships for HIV care and prevention programmes.” 

A basic care package incorporating IPT and screening for non-communicable diseases in Botswana

Of course, the basic care package can also refer to interventions that are more difficult to deliver in a kit, but which should nonetheless be provided for any person living with HIV. Once ART has been stably delivered, many of the health risks confronting people living on long term ART change, and according to a team from the Botswana-Baylor Children's Clinical Centre of Excellence (COE) in Gaborone, Botswana, there is a need to adapt and implement basic care packages in this setting as well.⁴

“Public policy on HIV in resource-limited settings has focused primarily on expanding access to antiretroviral therapy (ART), with less emphasis on preventive care for HIV-infected individuals. However, as life expectancy of HIV-infected patients improves due to ART and the importance of associated co-morbidities and chronic diseases increases, preventive care will become increasingly important,” the poster’s authors wrote.

After a comprehensive regional literature search, an assessment of clinic-specific prevalence data of other chronic conditions in their patients, national guidelines, and an assessment of public-sector resources, the team adapted the existing general basic care package to recommend, providing the following for adults enrolled in family model care at the facility:

Basic care package components: Botswana-Baylor Children’s Clinical Centre of Excellence

• Clean water provision and water storage

• Isoniazid Prophylaxis (IPT)

• Contraceptive provision

• Screening and treatment for depression, cervical cancer, breast cancer, hypertension, hyperlipidaemia, diabetes, viral hepatitis, sexually transmitted infections

• Assessment of other risks including: domestic violence, road safety, smoking and alcohol abuse.

“This preventive care package addresses the comprehensive health needs of stable HIV-infected adults in a particular resource-limited setting in an evidence-based, systematic manner, while minimizing manpower and resource costs of development,” wrote the authors, who suggested the same process be utilized to develop similar guidelines in other resource-limited settings.

It should be noted that another poster from the same facility reported on excellent uptake, retention in care, and completion of treatment of IPT at their site— in contrast to the poor retention and completion rates reported for the national programme in Botswana — which perhaps illustrates that the Botswana-Baylor Children's Clinical Centre of Excellence is indeed just that — and somewhat better resourced than other public sector facilities. Other presentations on TB and HIV integration are described in the HIV and TB in Practice section of this edition.

Leveraging HIV programmes to support diabetes services through integrated care, the team in Botswana concluded that that their basic care package should include screening and management for non-communicable diseases (NCDs), such as diabetes and hypertension, because there was a significant risk of these chronic illnesses in their population. Note, that this may be the mark of a more mature ART programme. In early ART programmes, where ART is initiated in late stage disease, other complications such as TB and opportunistic infections are initially more common].

Adoption of HIV care models to improve diabetes care

Another study, presented by Dr Miriam Rabkin of ICAP and Columbia University looked at the connection between HIV and NCDs from another angle — what can HIV platforms do to improve the health of people living without HIV but with other chronic illness, in this case, diabetes?⁵

“This work focused on… observations that in many resource-limited countries, the HIV scale up has created the first large scale chronic disease programmes; that a lot of the effort that’s put into implementation of HIV services actually goes to creation of continuity care; and that in many countries there are no other chronic disease programmes,” said Dr Rabkin, who pointed out that in these same countries, the burden of other chronic diseases — NCDs such as diabetes — is high and growing rapidly.

Notably, in Ethiopia, the prevalence of diabetes is estimated to be 2.0% compared to an HIV prevalence of about 2.1%. Ethiopia’s HIV programme receives more that US $1 billion in donor support and over 246,000 people have been put onto ART. But there is no diabetes programme, and little donor funding to support such programmes — in fact, in 2007, only 2.3% of the official development assistance went to all NCDs combined. Therefore, access to prevention, care and treatment services for diabetes remain out of reach for most people.

So the ICAP team conducted a rapid proof-of-concept study to explore the feasibility of adapting HIV-specific resources that were developed for an HIV clinic for use in a diabetes programme at Adama Hospital, a large urban hospital in Ethiopia. Following a baseline assessment, key strategies, systems and tools originally developed for the Adama HIV clinic, and so had already been tested, validated, translated, adopted and in use in this hospital, were quickly adapted and introduced into the outpatient department (OPD) where diabetic patients were being treated.

“It’s important to note that we didn’t introduce any new or experimental services into this study,” said Dr Rabkin. “We were following regional guidelines and local best practices for the care of diabetes. We didn’t create a new diabetes clinic, where there hadn’t been one there before, although we did steer follow-up visits to a specific day of the week.  No additional support was provided for medications, labs or transport. And no new staff were engaged for implementation.

Diabetes care was simply delivered using the strategies and systems (such as defaulter tracing, clinical mentorship, the implementation of peer educators programmes), registers, materials and types of job aids that had worked for the HIV programme:

Intervention Package

To see what sort of impact this would have, a chart review, knowledge, attitude and practice survey with the clinicians and focus groups were conducted with patients at baseline; then a follow up assessment was performed after six months.

Results

About 260 adult diabetic patients were actively engaged in care in the last 3 months. The median age was 47; about half were men; 57% were on insulin and they were a relatively sick group – 21% of them had been hospitalised for diabetes; 20% of them had been diagnosed with peripheral neuropathy; 8% had visual impairment which was attributed to diabetes but only 2% had had amputations.

At baseline, there were no algorithms or standard operating protocols in use to support diabetes care. Almost half of clinicians had completed their training over 10 years ago, and only 16% had received further training in diabetes management since graduating.

The team had identified a limited number of monitoring and evaluation indicators to track the performance of quality diabetes management. Key clinical services were rarely documented at baseline, but rose sharply after the intervention package was implemented.

“We saw a dramatic improvement in the documentation of service delivery,” said Dr Rabkin. “Now as all of you know, documentation does not put anything into delivery; but certainly if you’re providing chronic care and you can see at baseline only 2% of the patients had a weight documented - it’s really difficult to provide chronic care for a diabetic without knowing what that weight was before.”

She reiterated that this was done with no additional staff, but there was an expansion of services to include some things that hadn’t been there before including expanded peer educator services and point-of-service diabetes screening for patient’s family members.

Dr Rabkin pointed out that this was a proof-of-concept study — it was performed at a hospital, at the regional healthcare level, and not at the level of primary healthcare, which would be the point of care for most people.

“I think it does suggest some different approaches/different ways to think about integration, even if it’s not at the point of service yet. It’s a provocative idea that we’re going to follow up on,” she concluded.  

Practical take home messages

During the poster discussion session on integration, presenters were asked what advice they had to offer to those attempting to do this kind of cross-discipline and  cross-systems integration.

“My advice would be to evaluate and see what is working, and be willing to keep changing direction, as you find things that are working and aren’t working – especially when you’re out there in the community,” said Dr Sue Ann Meehan, who presented a study on integrating TB screening at HIV counselling and testing facilities.

“I agree, the most important thing is design something which is tailored to your situation,” said Dr Sabine Hermans, who presented on a one stop clinic for TB and HIV services. She added that engaging everyone who was going to work in their clinic was very important “Trying to design something and make other people implement it, wouldn’t work. It is the whole team effort, which made it work, and which is continuously adapting it.”

Simon Sensalire stressed the importance of managing “opportunistic infections, where they are not controlled — and provide opportunities for people living with HIV by way of  providing for these interfaces, so that you really make treatment viable,” he said. 

“I would say to be extremely precise about what questions you are asking regarding the integration of services: What is the purpose of integrating these services? Because integration means different things to different people and there’s a lot of confusion about what we’re still seeking to integrate with what, and why?” said Dr Rabkin. “So some of these studies are designed for the purpose of improving the quality of care, for people living with HIV. In others we have heard about at this conference – the questions that are being asked about integration include questions of efficiency; they include questions of cost; and they also include questions about the impact of HIV scale-up on health systems.

“We need to shift our mindset,” said Dr Ocero. “If we want to provide an integrated service and improve the quality of care for our patients, then we have to look at it from a broader sense. Therefore the health system has to also think of integration – those that coordinate the health system also have to think in a more integrated manner.”

Dr Ocero also pointed out that many of these presentation concerned larger facilities, but that at smaller facilities where services are being decentralised, the same health workers are already providing multiple integrated services.

One audience member asked how to avoid healthcare workers feeling as though integration was just adding another task to their already heavy burden of work — and how to give them more incentive to take on the new task.

“In our clinic, we operate as more of a team – than the rest of the HIV clinic. So people really like to work in our clinic because they get more support, there’s also more opportunities to be supervised and also to grow career options,” said Dr Hermans.

Mr Sensalire agreed that in his setting, the people who operate at the community level are not service specialists in one area, but are trained in multiple disciplines. “They go out into the community and mobilize and sustain the skills that the community needs.”     

Dr Rabkin noted that these skills have to evolve in HIV caregivers depending upon the changing needs of their patients.

“We’ve seen this very issue amongst HIV caretakers in wealthy countries, as we evolve from being sort of ART delivery people, to people who are really spending a lot of time speaking about hyperlipidaemia and diabetes and cardiovascular disease, and so I think some of the lessons can be extrapolated and there’s a whole lot to do with multidisciplinary teams and task shifting,” she said.

Dr Ocero stressed the importance of the community in the delivery of integrated services. “Integrating our services into the community is really synonymous with trying to enhance the continuum of care.  And with NUMAT, and many others, we show that we first ‘attempt’ to work with PLHIV teams – that has been something that we have been trying to work upon. Previously we were working with ‘Expert Clients’ – people who are self-motivated, who were trained and given basic training and became people who could support the health workers; go out into the communities, enhance the care and retention – you know, basic psychosocial support – and navigate the patients to other forms of care that they require. And in this sense, integration was still continuing in that direction.”

Finally, a programme director in the audience said that with all the pressure to ‘integrate’ “we have to know the right question – what are we asking? Because integration can increase efficiency but decrease quality of care,” she said. She suggested that it might be worthwhile to consider if the integration of certain services is AFASS — the old acronym used for determining whether or not the provision of formula food was advisable in a setting based upon if it was acceptable, feasible, affordable, sustainable and safe in a particular setting.

Dr Rabkin agreed that that was an interesting approach. She pointed out that most of the examples at the conference involved the integration of clinical services, but there could also be integration around each of those six health systems building blocks, around procurement systems financing, governance and supervision.

“There are opportunities for integration, and there are also risk with integration, and I think a lot of these questions have not necessarily been answered. Its worth considering whether what you integrate is connected to your output. What’s your rationale? Are you trying to be cheaper – which I think a lot of the pressure towards integration is about efficiency. Okay, lets prove it’s more efficient. It sounds like it should be but I don’t’ think there’s the data necessarily in all of our contexts. Is it better? It might be better; it seems better. I would prefer the patient to go to a one-stop shop rather than having to go to all different service points. But then again, there is aa research agenda needed around integration.”