Pr Jean-Paul Moatti Chair of ANRS Scientific Committee on Public Health & Social Sciences

Evaluation of access to ART and decentralization of health care

delivery in Cameroon

French Agency for AIDS Research (ANRS) Program in Economic & Social Sciences

Dr. Fred Eboko, IRD UMR 912 INSERM-IRD U2 Marseille

Pr Jean-Paul MoattiChair of ANRS Scientific Committee on Public Health & Social

Sciences

Ministère de la Santé Publique du Cameroun

FPAE CASS-RT / Un. Ydé I GRAPS / Un. Ydé II

National ART Programme in Cameroon

Use of preexisting decentralized framework of health care system (<1992- 174 districts) for ART delivery:– In 2001- 2002: from reference centers in central hospitals to provincial

hospitals (24 ATCs)– From 2005: 106 MUs in district hospitals (WHO public health approach

for care algoithms) including 35 from private sector

Generic oriented procurement of ARV drugs (70% total) through monopoly of imports for CENAME (National Agency for Drug Procurement)– Decrease of monthly ART prices for patients: from 250,000 FCFA in

2000 to between 3,000 and 7,000 FCFA in 2004 (1$= 496.6 FCFA)– Gratuity of ARVs introduced in May 2007

85% of total AIDS budget (139,2 Million US$-2004/2007) funded by foreign aid

Guidelines for ART decentralization1

Initial evaluation of diagnosed HIV patients: physical examination + CD4 count or complete cell blood count (CBC) when CD4 counter not availableFor patients eligible for ART: pre-therapeutic check-up including CBC at the district level; CBC and CD4 count at the other levels or when availableEvaluation of ART eligibility using the WHO classification (2005) when CD4 count not available: - WHO stage III or IV and WHO stage II when Total Lymphocytes<1200 c/mm3

Collegial decision by the therapeutic committee about ART protocols 4 first line regimens available: 2 NRTI + 1 NNRTI

1 National guidelines for the district level, 2005

Rapid national scale up of access to ART in Cameroon

Year Nb of ART Facilities

Nb of ART-treated (adults) (%rate of coverage)

2001 18 600

2003 23 9,000

2005 89 17,940 (22%)

2007 109 45,817 (53%)

2008june

132 53,238 (58%)

Independent evaluation of national ART program in CameroonRequested by Ministry of Public Health of

Cameroon

Carried out by Universities of Yaoundé and ANRS research teams Evaluate an ongoing process and propose recommendations for

improvement Cross-fertilization of quantitative and qualitative methods

Included 4 research projects :– Decentralization of ARV access in Africa: Evaluation of the treatment of

patients on ARV in district hospitals using a streamlined follow-up approach (STRATALL)

– Impact of the Cameroonian access to ARV program on the treatment and living conditions of the HIV infected population (EVAL)

– The problem of access to ART in Cameroon. Political Issues, Advances, Limits and Perspectives of decentralization of health care (POLART)

– Scaling up and procurement of drugs and biological monitoring tools (CEPN)

Objectives

Evaluation of the Impact of access to ART on the living conditions of PLWHA according to levels of care delivery

Efficiency Equity Democratization

Evaluation of the impact on the health system– Impact on medical knowledge and practice– Changes introduced in the organization of health care– Institutional impact on decentralization of health care

delivery

Data collection between September 2006 & March 2007

EVAL ANRS 12 116 Pr. Moatti (Inserm Marseille), Pr. Abega (UCAC Yaoundé)

Cross-sectional survey in a random sample of 3,151 adults, HIV diagnosed for at least 3 months and seeking care in 14 ATCs & 13 MUs in 6 provinces (response rate = 90%)

Survey in the exhaustive sample of HIV care physicians in the same centers (n=97, resp. rate= 92%) and stratified sample of other healthcare personnel (n= 208, resp.rate= 82%)

Data collection on characteristics of the 20 public and 7 private health facilities

Semi-structured interviews (n=25 health personnel & 53 patients)

EVAL ANRS 12 116 (methods)

Characteristics of the 27 ART-delivery centers in the EVAL Survey

Availability of equipment = complete cell blood count, CD4 cell count, transaminases, glycemia, creatinemia, amylasemia, pregnancy test, viral load, triglycerides and cholesterol

Median(IQR)

Central(n=8)

Province(n=6)

District(n=13)

Pval.

Nb beds 234 (120-300)

164 (37-230)

120 (93-166)

0.19

Level Equipment

9.0 (8.2-9.0)

9.0 (8.5-9.0)

7.0 (4.5-8.0)

0.004

CD4 cell count 7 6 7 0.06

FTE Phys. 4(3-7)

2(2-4)

3(2-3)

0.028

FTE Total 18(16-21)

13(10-18)

12(8-15)

0.018

Nb HIV+ pts 699(299-2608)

732 (421-1166)

150 (83-441)

0.001

Nb ART Initiation/mth

50 (34-114)

43(28-57)

15(11-28)

0.002

Nb HIV+/phys 211(126-514)

335(129-797)

61(35-164)

0.009

EVAL Physicians’ surveyNo significant differences according to the level of decentralization in terms of (n=97):

N (%) or median [IQR]Good knowledge of national protocols - right answers to >=5 in 6 questions on national protocols 61 (62.9%)Good knowledge of criteria of ART eligibility- right answers to >=4 in 5 questions on criteria of ART eligibility 74 (76.3%)Knowledge on ART management- score ranging from 0 to 35 points 28 [23; 30]Number of years of experience in PLWHA care 4.0 [2.0; 7.0]

Employment status : - in public hospitals : civil servant 60 (76.9%)- in private hospitals : contractual 12 (63.2%)

Monthly income perceived from the hospital x 103 FCFA 250 [200; 300]Monthly income considered as a fair remuneration x 103 FCFA 400 [300; 600]

Not at all or rather not satisfied with the income perceived 74 (76.3%)

1$= 496.6 FCFA at the time of the survey

Physicians knowledge and experience

Working conditions

EVAL Physicians’ survey

But some significant differences in terms of practices and opinions on the ART policy implementation (n=97)

Central level(N=40)

Provincial level (N=22)

District level (N=35)

P-value

Practices

Participation to the therapeutic committee: - At each meeting or almost 15 (37.5%) 10 (45.5%) 17 (48.6%) 0.33

Task shifting in consultation: - yes 15 (37.5%) 8 (36.4%) 18 (51.4%) 0.01

Opinions on the ART policy implementationWorkload: - too heavy 18 (45.0%) 5 (22.7%) 7 (20%) 0.04

Perception of policy implementation: - Score ranging from 0 to 21 - Median [IQR] 11 [9;13] 10 [9;12;5] 13 [10;15] 0.04

Disagreements with decentralisation policy : - Inadequacy of technical means- Inadequacy of supervision- Inadequacy of decisional autonomy

25 (62.5%)29 (72.5%)21 (52.5%)

16 (72.7%)16 (72.7%)14 (63.6%)

15 (42.9%)11 (31.4%)15 (42.9%)

0.060.010.31

EVAL Physicians’ survey - qualitative data

Structural constraints at the three levels of decentralization:Poor working conditions – Lack of equipments and frequent breakdowns– Low wages and insecure employment / status

Generalized dissatisfaction and demotivation

Patients’ poverty– Incapacity of patients to pay for prescribed treatments and

recommended biological tests

ART supply deficiency: shortage

Lack of appropriate HR qualification, especially for psychological care

EVAL Physicians’ survey - qualitative data

Organizational constraintsA doctor-intensive policy- No definition in the national policy of a task shifting strategy

and procedures Large physicians’ workloads and insufficient time per patients Or conversely: unorganized and high task-shifting

Involvement of Community Health Workers (CHW’s) without a clear definition of their roles

Conflicts of rolesTensions between healthcare workersExclusion and frustration Desire to move from HIV-services and to give-up the profession

Characteristics of HIV-infected patients in the EVAL Survey (n=3,151)

Central(n=1112)

Province(n=1017)

District(n=1022)

Pval.

Female 70.8% 71.5% 70.9% 0.93

Mean Age (sd) 37.9(9.2)

38.0 (9.2)

36.9 (9.4)

0.012

Edu>Primary 72.3% 51.5% 39.3% 0.001

Living in couple

52.8% 43.0% 47.1% 0.001

<Poverty line 65.9% 76.8% 82.5% 0.001

Informal sector 40.4% 54.1% 61.1% 0.001

ART-treated 78.0% 83.5% 73.4% 0.02

Eligible Non ART treated

9.0% 5.9% 10.3% 0.001

Characteristics of ART-treated patients (>6months) in the EVAL Survey (n=2,132)

< 1hr Central(n=760)

Province(n=761)

District(n=611)

Pval.

Mths <HIV diagnosis

22.9 24.9 16.1 0.001

Triomune ART regimen

52.9% 80.9% 86.1% 0.001

CD4<200 21.1% 21.5% 21.9% 0.82

Highadherence4 wks

44.5% 58.1% 61.2% 0.001

ARV shortage 3mths

14.1% 11.7% 4.4% 0.001

1st visit <1mth after HIV diag

56.6% 56.2% 64.3% 0.001

Catastrophic Hlth Exp

42.1% 43.5% 46.3% 0.23

Waiting time <1hour

43.0% 83.7% 95.1% 0.001

4 OUTCOME VARIABLES- average monthly gain in CD4 cells/mm3 since initiation of

treatment, - adherence to ART in previous 4 weeks (high vs

moderate/low), - physical and mental HRQL (MOS-SF12)

Two-level models (mixed effects regression) for hierarchically structured data (patients nested within care centres)

All variables at p<0.2 in univariate two-level analysis initially introduced in the multivariate model

Multivariate statistical analysis (EVAL-patients’ survey

Multilevel mixed effects models (ref= central level of care)

Coef/ IC 95%

Provincial P val District P val

CD4 Gain/mth

-0.27(-049/-0.04)

0.02 -014(-035/0.07)

0.19

High Adherence

2.19(1.03-4.68)

0.04 1.97(1.03-3.77)

0.04

Phys HRQL

0.09 (-017:0.34)

0.50 -0.03(-0.25/0.19)

0.77

Mental HRQL

0.19(-0.20/0.58)

0.34 0.34(0.00/0.69)

0.05

Eval Survey- qualitative interviews of managers and health

professionals

Decentralization can come in a variety of forms: deconcentration, devolution, privatization

Decentralization of access to ARV in Cameroon corresponds in a general way to a process of deconcentration

Trend toward recentralization of drug procurement supply chain

Problems of referral between levels of care

Growing tensions between physicians involved in HIV care and colleagues

Perceived inequity between HIV and other diseases

Main lesson of the EVAL study

Decentralization of ART-delivery is clinically feasible and brings additional benefits (more equal access to ART for the poor, better mental quality of life, more adherence)

Potential negative impact on decentralization of health system if “verticalization” is pursued without more integration in global reform for

Human resource crisis Health financing Procurement of drugs

Issues for the future of the Cameroonian program

Long term and free financial sustainability of access to medicines? Optimal degree of decentralization to enable scaling-up?

New distribution of tasks between healthcare providers (“task shifting”) to find solutions to the Human Resources crisis?

Impact of AIDS program on the fight against other diseases (tuberculosis, malaria) and on the global reinforcement of the health care system?