May, 2017 - SUPPORT Summary of a systematic review | print this article |

Does decentralised HIV treatment improve health outcomes?

Many people living with HIV who need antiretroviral therapy are unable to access or remain in care. This is often because of the time and cost required to travel to health centres. One strategy to address this problem is to move antiretroviral delivery from hospitals to more peripheral health facilities or even beyond health facilities. This could increase the number of people with access to care, enhance retention in treatment programmes, improve health outcomes and reduce costs to people living with HIV and AIDS and health services. However, there are some concerns about the quality of decentralised care and whether health outcomes are equivalent to more centralised care.

Key messages

  • Partial decentralisation of HIV treatment (starting care at hospital and then moving to health centre care) probably reduces the combined number of people who die or are lost to care at one year, and may reduce the costs of travel for patients.
  • Full decentralisation of HIV treatment (starting and continuing care at a health centre) probably reduces the number of people lost to care but it is uncertain if it reduces deaths at one year.
  • Decentralisation of HIV treatment from facility to community probably leads to little or no difference in the number of people who die or are lost to care at one year.
  • Decentralisation of HIV treatment from facility to community may reduce total costs to people living with HIV and AIDS and to the health service.
  • Most of the included studies were conducted in low-income countries.

Background

Although there has been considerable progress in improving access to antiretroviral therapy (ART) to date, global coverage for ART is still around 50% of those eligible for treatment, and 25% of people on treatment are not retained in care within 24 months of initiating ART. Decentralisation of ART care delivery from hospitals to more peripheral health facilities is an important strategy for addressing these problems. Decentralisation of care broadly means relocating services from centralised sites (i.e. hospitals) to peripheral health centres or lower levels of healthcare, generally geographically closer to the homes of people living with HIV and AIDS. Three types of decentralisation can be considered:

• Partial decentralisation: starting ART at the hospital, then moving to a health centre to continue treatment.

• Full decentralisation: starting and continuing ART at a health centre.

• Decentralisation from facility to community: ART is started at a health centre or hospital and thereafter provided in the community. Support for treatment may be provided by a family member, a lay or community health worker or through outreach by a health worker based in a primary healthcare clinic.

 



About the systematic review underlying this summary

Review objectives: To assess the effects of decentralised HIV care in relation to initiation and maintenance of antiretroviral therapy.
Type of What the review authors searched for What the review authors found
Study designs & interventions Randomised and non-randomised trials, controlled before-after studies and well-designed cohort studies assessing any form of decentralised care delivery model for the initiation of ART, continuation of ART, or both.

16 included studies: two cluster trials, two prospective cohorts and 12 retrospective cohort studies.

 

The studies examined partial decentralisation (6 studies), full decentralisation (7), and decentralisation from facility to community (3).


Participants HIV-infected patients at the point of initiating treatment, and patients already on treatment requiring maintenance and follow-up.
HIV infected patients. Three included children only, two included adults and children and the rest included adults only.
Settings Community, health centre and hospital settings.
Studies from rural and urban areas in South Africa (4 studies), Malawi (3 studies), Ethiopia (2), Uganda (2), Kenya (1), Swaziland (1), and Thailand (1). One study examined data from five countries in Africa (Kenya, Lesotho, Mozambique, Rwanda and Tanzania).
Outcomes

Primary: Lost to care at one year, death, and a composite outcome of both.

Secondary: Time to starting antiretroviral therapy, new diagnoses of tuberculosis co-infection, virologic and immunologic response to ART, new AIDS-defining illness, patient satisfaction with care, and cost to the provider.


ll primary outcomes, virologic and immunologic response to ART, costs to people living with HIV and AIDS and costs to the health service, and patient satisfaction with care.
Date of most recent search:March 2013
Limitations:This is well-conducted systematic review with only minor limitations.

Kredo T, Ford N, Adeniyi FB, Garner P. Decentralising HIV treatment in lower- and middle-income countries. The Cochrane database of systematic reviews. 2013;6: CD009987

Summary of findings

16 studies were included in the review. All studies evaluated decentralisation of care and eight also evaluated task shifting from doctors to other types of healthcare providers. Three studies examined treatment in children only, two included adults and children and the rest included adults only.

1) Partial decentralisation: starting ART at the hospital, then moving to a health centre to continue treatment 

Four studies considered this option:

  • Partial decentralisation probably reduces the combined number of people who die or are lost to care at one year. The certainty of this evidence is moderate.
  • Partial decentralisation may reduce the number of people who are lost to care at one year. The certainty of this evidence is low.
  • Partial decentralisation may reduce death at one year. The certainty of this evidence is low.
  • Partial decentralisation may reduce the cost of travel. The certainty of this evidence is low.

Effect of partial decentralisation compared to usual care on patient outcomes

People              HIV patients

Settings            Community, health centre and hospital settings

Intervention     Partial decentralisation

Comparison      Usual care

Outcomes

Usual care

Partial

decentralisation

Relative effect

(95% CI)

Certainty

of the evidence

(GRADE)

Absolute effect (95% CI)

Death or lost to care (combined)

Follow-up: 12 months

218 per 1000

100 per 1000

(63 to 155)

RR 0.46

(0.29 to 0.71)

Moderate

Lost to care

Follow-up: 12 months

134 per 1000

74 per 1000

(60 to 93)

RR 0.55

(0.45 to 0.69)

Low

Death

Follow-up: 12 months

84 per 1000

28 per 1000

(11 to 73)

RR 0.34

(0.13 to 0.87)

Low

Cost of travel

Mean 1.5 USD

Mean 0.74 USD

-

Low

Margin of error = Confidence interval (95% CI) RR: Risk ratio USD: United States Dollar

GRADE: GRADE Working Group grades of evidence (see above and last page)

2) Full decentralisation: starting and continuing ART at a health centre

Four studies considered this option:

  • It is uncertain if full decentralisation reduces the number of people who die or are lost to care at one year as the certainty of this evidence is very low.
  • Full decentralisation probably reduces the number of people who are lost to care at one year. The certainty of this evidence is moderate.
  • It is uncertain if full decentralisation reduces deaths at one year as the certainty of this evidence is very low.

Effect of full decentralisation compared to usual care on patient outcomes

People               HIV patients

Settings             Community, health centre and hospital settings

Intervention      Full decentralisation

Comparison       Usual care

Outcomes

Usual care

Full decentralisation

Relative effect

(95% CI)

Certainty

of the evidence

(GRADE)

Absolute effect (95% CI)

Death or lost to care

Follow-up: 12 months

365 per 1000

256 per 1000

(172 to 373)

RR 0.7

(0.47 to 1.02)

 

Very low

Lost to care

Follow-up: 12 months

270 per 1000

81 per 1000

(46 to 146)

RR 0.3

(0.17 to 0.54)

Moderate

Death

Follow-up: 12 months

97 per 1000

106 per 1000

(61 to 185)

RR 1.1

(0.63 to 1.92)

 

Very low

Margin of error = Confidence interval (95% CI) RR: Risk ratio GRADE: GRADE Working Group grades of evidence (see above and last page)


3) Decentralisation from facility to community: ART is started at a health centre or hospital and thereafter provided in the community

Two studies in adult populations from Kenya and Uganda considered this option.

Decentralisation from facility to community:

  • Probably leads to little or no difference in the number of people who die or are lost to care at one year. The certainty of this evidence is moderate.
  • Probably leads to little or no difference in the number of people who are lost to care at one year. The certainty of this evidence is moderate.
  • Probably leads to little or no difference in deaths at one year. The certainty of this evidence is moderate.
  • May reduce total costs to people living with HIV and AIDS and to the health service. The certainty of this evidence is low.

Effect of decentralisation from facility to community compared to usual care on patient outcomes

People               People living with HIV and AIDS

Settings             Community, health centre and hospital settings

Intervention      Decentralisation from facility to community

Comparison        Usual care

Outcomes

Usual care

Decentralisation from facility to community

Relative effect

(95% CI)

Certainty

of the evidence

(GRADE)

Absolute effect (95% CI)

Death or lost to care

Follow-up: 12 months

106 per 1000

101 per 1000

(66 to 155)

RR 0.95

(0.62 to 1.46)

Moderate

Lost to care

Follow-up: 12 months

26 per 1000

21 per 1000

(8 to 57)

RR 0.81

(0.3 to 2.21)

Moderate

Death

Follow-up: 12 months

55 per 1000

57 per 1000

(35 to 91)

RR 1.03

(0.64 to 1.65)

Moderate

Total cost to people living with HIV and AIDS*

USD 54/year

USD 18/year

-

Low

Costs to the health service^

USD 838 / year /

patient

USD 793 / year /

patient

-

Low

Margin of error = Confidence interval (95% CI) RR: Risk ratio GRADE: GRADE Working Group grades of evidence (see above and last page)

*Transport, lunch, childcare costs, lost work time. Based on data from one study.

^ Staff, transport, drugs, laboratory, training, supervision, capital and utilities costs. Based on data from one study.

Relevance of the review for low-income countries

Findings Interpretation*
APPLICABILITY
14 of the studies included in the systematic review were conducted in low -income countries and 2 in middle-income countries.

All but one of the included studies were from Africa. The applicability of the findings to other low-income settings is therefore unclear.

The findings may be applicable in settings where a reasonable infrastructure exists for the decentralisation of HIV treatment. This needs to include facilities, referral systems, human resources and supplies.

In some countries, obstacles to task shifting or decentralization include regulations governing the work scope of different health workers and the views of labour unions representing health workers.

The acceptability of decentralisation to people living with HIV and AIDS and to healthcare providers needs to be considered. Service planners also need to consider the impacts of decentralisation on total costs for both people living with HIV and AIDS and the health service.


EQUITY
There was no information in the included studies regarding the differential effects of the interventions on resource-disadvantaged populations.

The resources needed to support decentralised care, and to ensure appropriate referral between levels of care, may be less available in disadvantaged settings.

Decentralising care from facilities to the community may improve access to care and outcomes for disadvantaged groups, and thereby improve equity, through reducing total costs to people living with HIV and AIDS and reducing the number of people lost to care.


ECONOMIC CONSIDERATIONS
The systematic review found that decentralisation may reduce total costs to people living with HIV and AIDS and to the health services.

Little data on costs were available for different decentralisation options.

Different models of decentralising HIV treatment may have different cost impacts for people living with HIV and AIDS and for health services. Care needs to be taken that the costs to individuals are not increased through, for example, higher travel costs or user fees. Local costing studies may be desirable before scaling up these interventions.

Decentralisation may lead to changes in the use of healthcare provider time, supplies and laboratory tests at peripheral health facilities, with implications for other services delivered at these sites and for local budgets.

The longer term economic consequences of decentralisation are not clear and need to be monitored.


MONITORING & EVALUATION
Two cluster trials, two prospective cohorts and 12 retrospective cohorts contributed data to this review.

Large pragmatic trials may be helpful in evaluating some decentralisation options. Where decentralisation is implemented at scale, monitoring may be needed of costs and of impacts on workload, support needs and supply chains at peripheral facilities.


*Judgements made by the authors of this summary, not necessarily those of the review authors, based on the findings of the review and consultation with researchers and policymakers in low-income countries. For additional details about how these judgements were made see: www.supportsummaries.org/methods


 

Additional information

Related literature

Brinkhof MW, Pujades-Rodriguez M, Egger M. Mortality of patients lost to follow-up in antiretroviral treatment programmes in resource-limited settings: systematic review and meta-analysis. PloS one. 2009;4(6):e5790.

 

Callaghan M, Ford N, Schneider H. A systematic review of task- shifting for HIV treatment and care in Africa. Human Resources for Health. 2010;8:8.

 

Denno DM, Chandra-Mouli V, Osman M. Reaching youth with out-of-facility HIV and reproductive health services: a systematic review. The Journal of Adolescent Health: official publication of the Society for Adolescent Medicine. 2012;51(2):106-21.

 

Young T, Busgeeth K. Home-based care for reducing morbidity and mortality in people infected with HIV/AIDS. Cochrane Database Syst Rev. 2010(1):CD005417

 

This summary was prepared by

Agustín Ciapponi, Instituto de Efectividad Clínica y Sanitaria, Buenos Aires, Argentina

 

Conflict of interest

None declared. For details, see: www.supportsummaries.org/coi

 

Acknowledgements

This summary has been peer reviewed by: Tamara Kredo and Mike Callaghan.

 

This review should be cited as

Kredo T, Ford N, Adeniyi FB, Garner P. Decentralising HIV treatment in lower- and middle-income countries. The Cochrane database of systematic reviews. 2013;6:CD009987.

 

The summary should be cited as

Ciapponi A, Does decentralised HIV treatment improve health outcomes? A SUPPORT Summary of a systematic review. May 2017. www.supportsummaries.org

 

Keywords

evidence-informed health policy, evidence-based, systematic review, health systems research, health care, low and middle-income countries, developing countries, primary health care.

decentralisation, down referral, delivery of health care, health services, accessibility, HIV, retention in care, antiretroviral therapy.

 



Comments