Journal of Clinical Research In HIV AIDS And Prevention

Journal of Clinical Research In HIV AIDS And Prevention

Current Issue Volume No: 3 Issue No: 2

Research-article Article Open Access
  • Available online freely Peer Reviewed

  • Evaluation Of An External Quality Assessment Program For HIV Testing In Tigray North Ethiopia 2016

    1 Tigray health research institute, Mekelle 

    Abstract

    Background:

    Point-of-care diagnostic tests (POCTs) are increasingly used in both developing and developed countries. They allow same day testing and treatment at remote locations where no laboratory support is available. Quality control measures, which are routinely used in laboratories, have not been widely implemented for POCTs. This aimed to assess the integrity of the entire laboratory testing process, and aims to educate and improve performance in quality of HIV rapid testing.

     Methods:

    A health facility based cross section study was conducted from April to June 2016.Randomly selected health facilities were participated in the external quality assessment. Onsite evaluation and panel test were used to collect data using structured checklists and formats. Data was entered and analyzed using SPSS version 16.

     Results:

    Between April to June 2016, a total of 60 health facilities (145 testing points) from governmental health facilities (hospitals and health centers) were participated in the study. Among the participated testing points 41% have no designated area, 40% have no clean water for hand washing and 51% have no national test algorithm. The average performance of testing points was varies from 89.6% to 99.1% (Laboratory 99.1%, ANC 90.4%, TB clinic 91.4% and VCT 89.6%). In a multivariable logistic regression model, didn t follow national testing algorithm to report client test results have statistical significance.

     Conclusions:

    High quality test results underpin accurate diagnosis and appropriate treatment for patients. But in the study area the score of proficiency testing result and coverage of training is slightly low comparing to other findings. Therefore following national testing algorithm to report client test results, training and monitoring are critical points to improve the proficiency testing score of testing points.

    Author Contributions
    Received Oct 03, 2016     Accepted Dec 12, 2016     Published Nov 21, 2017

    Copyright© 2017 Tesfahuneygn Gebrehiwet, et al.
    License
    Creative Commons License   This work is licensed under a Creative Commons Attribution 4.0 International License. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

    Competing interests

    The authors have declared that no competing interests exist.

    Funding Interests:

    Citation:

    Tesfahuneygn Gebrehiwet, Gebreegziabher Gebremichael (2017) Evaluation Of An External Quality Assessment Program For HIV Testing In Tigray North Ethiopia 2016 Journal of Clinical Research In HIV AIDS And Prevention. - 3(2):1-8
    DOI 10.14302/issn.2324-7339.jcrhap-17-1679

    Results

    Result

    A total of 145 testing points from governmental health facilities (hospitals and health centers) participated in the study. Among the participated testing points 41% have no designated area, 40% have no clean water for hand washing and 51% have no national test algorithm (Table 1).

    General back ground characteristics of testing points in HIV testing
    characteristics Yes N (%) No N (%)
    Designated area for HIV testing 86 (59) 59 (41)
    Availability of clean water and soap for hand washing 87 (60) 58 (40)
    Availability of national test algorithm 94 (64.8) 51 (35.2)
    Sop for each HIV rapid test 72 (49.7) 73 (50.3)
    Kits and supplies used within their expiry date 139 (96) 6 (4)
    IQC practice 65 (45) 80 (55)
    Training for HIV rapid test 81 (56) 64 (44)
    Clean and organized working area 132 (91) 13 (9)
    Appropriate disinfectants 110 (76) 35 (24)
    Waste segregation 101 (70) 44(30)
    kits stored according to manufacturer recommendation 137 (94.5) 8 (5.5)
    Inventory management 59 (41) 86 (59)
    job aids on specimen collection 52 (36) 93 (64)
    Availability of sufficient kits and supplies 122 (84) 23 (16)
    Labeling with client identification number 12 (89) 16 (11)
    Availability of timer 44 (30) 101 (70)
    testing procedures adequately followed 108 (74.5) 37 (25.5)
    Following the national testing algorithm to report the client test result 128 (88) 17 (12)
    PT Result of Testing Points

    Most (82.8%) of the testing points score above the expected standard. From the total testing points 118 (81.4%) testing points score 100%, 6 (4.2%) testing points score 33%, 19 (13%) testing points score 67% and 2 (1.4%) testing points score 83%. The average performance of testing points was varies from 89.6% to 99.1(Laboratory 99.1%, ANC 90.4%, TB 91.4% and VCT 89.6%).

     Factors Associated with Poor HIV Rapid test PT Performance

    In a univariable logistic regression analysis, have no national algorithm in testing sites has high risk (odd ratio 2.86) to report false client results than having national test algorithm at P value (0.03). Using timer to report client result has reduced false HIV results six times than not using timer (P value 0.015). An adequately following test procedure during HIV testing has great advantage. Because it reduces false negative results 4 times than not adequately following test procedure (P value 0.002) (Table 2).

    Association of health service providers with HIV rapid test PT results
    Characteristics Response categories PT result Crude OR (95% CL) P-value
    Meet the standard Below the standard
    Testing points laboratory 36(97.3) 1(2.7) Reference  
    ANC 29(76.3) 9(23.7) 11.172(1.34, 93.37) .026
    TB 29(82.9) 6(17.1) 7.45(.85, 65.41) .070
    VCT 26(74.3) 9(25.7) 12.462(1.486, 104.51) .020
    Availability of national test algorithm Yes 83(88.3) 11(11.7) Reference  
    No 37(72.5) 14(27.5) 2.86(1.19,6.89) .019
    Sop for each HIV rapid test Yes 64(88.9) 8(11.1) Reference  
    No 56(76.7) 17(23.3) 2.43(.97,6.10) .057
    Kits and supplies used within their expiry date Yes 115(82.7) 24(17.3) Reference  
    No 5(83.3) 1(16.7) .96(.11,8.60) .97
    IQC practice Yes 59(90.8) 6(9.2) Reference  
    No 61(76.2) 19(23.8) 3.06(1.14,8.20) .026
    Training for HIV rapid test Yes 67(82.7) 14(17.3) Reference  
    No 53(82.6) 11(17.2) .99(.42,2.37) .99
    kits stored according to manufacturer recommendation Yes 115(83.9) 22(16.1)    
    No 5(62.5) 3(37.5) 3.14(.70,14.10) .14
    job aids on specimen collection Yes 48(92.3) 4(7.7) Reference  
    No 72(77.4) 21(22.6) 3.50(1.13,10.83) .030
    Labeling with client identification number Yes 110(85.3) 19(14.7) Reference  
    No 10(62.5) 6(37.5) 3.47(1.13,10.68) .030
    Availability of timer Yes 42(95.5) 2(4.5) Reference  
    No 78(77.2) 23(22.8) 6.17(1.39,27.55) .017
    testing procedures adequately followed Yes 96 12 Reference  
    No 24 13 4.3(1.8,10.7) .002
    Following the national testing algorithm to report the client test result Yes 111(86.7) 17(13.3) Reference  
    No 9(52.9) 8(47.1) 5.80(1.97,17.10) .001
    Factors associated with poor HIV rapid test PT performance
     characteristics Response categories PT result Crude OR (95% CL) Adjusted OR (95% CL) P-value
    Meet the standard Below the standard
    Testing points laboratory 36(97.3) 1(2.7) Reference Reference  
    ANC 29(76.3) 9(23.7) 11.172(1.34, 93.37) 3.58(.30,42.22) .311
    TB 29(82.9) 6(17.1) 7.45(.85, 65.41) 2.70(.20,35.05) .454
    VCT 26(74.3) 9(25.7) 12.46 (1.49, 104.51) 5.71(.46,71.01) .175
    Availability of national test algorithm Yes 83(88.3) 11(11.7) Reference Reference  
    No 37(72.5) 14(27.5) 2.86(1.19,6.89) 2.06(.69,6.22) .198
    IQC practice Yes 59(90.8) 6(9.2) Reference Reference  
    No 61(76.2) 19(23.8) 3.06(1.14,8.20) 1.80(.50,6.52) .374
    job aids on specimen collection Yes 48(92.3) 4(7.7) Reference Reference  
    No 72(77.4) 21(22.6) 3.50(1.13,10.83) 1.70(.38,7.68) .492
    Labeling with client identification number Yes 110(85.3) 19(14.7) Reference Reference  
    No 10(62.5) 6(37.5) 3.47(1.13,10.68) 2.05(.48,8.74) .332
    Availability of timer Yes 42(95.5) 2(4.5) Reference Reference  
    No 78(77.2) 23(22.8) 6.17(1.39,27.55) 3.17(.45,22.18) .246
    testing procedures adequately followed Yes 96 12 Reference Reference  
    No 24 13 4.3(1.8,10.7) 1.29 (.35,4.73) .701
    Following the national testing algorithm to report the client test result Yes 111(86.7) 17(13.3) Reference Reference  
    No 9(52.9) 8(47.1) 5.80(1.97,17.10) 11.57(2.46,54.35) .002

    In a multivariable logistic regression model, didn’t follow national testing algorithm to report client test results has statistical significance (Table 3).

    Discussion

    Discussion

    External quality assessments is a method used for assessing laboratory and health care professionals performance and allows for the evaluation of inter-laboratory proficiency tests and the identification of related problems. Accordingly, it affords grounds for corrective and preventive actions on a regular basis 10. Measures to control the quality of result in HIV diagnostic laboratories are extremely important, because of the consequence of either false Positive or false negative results are huge 9.

    This study demonstrates low (56%) coverage of HIV training in the testing points. This is different from a study conducted in south Ethiopia which was 85% 11. This difference may be due to difference in sample size and characteristics of study participants, our study participants were laboratories and other health professionals whom perform HIV rapid test in TB clinic, VCT, ANC but their study participants were only laboratory and their samples were only 20 health facilities. From the assessed testing points, 64 (44%) health professionals who conduct HIV testing were found to have no trained in HIV testing. This is incomparable with study conducted in Ethiopia, which reported only14% were found to have no trained in HIV testing 11. In this study 12% study participants didn t follow national test algorithm. This is in line with study conducted in Ethiopia, which demonstrates 10% of the study participants didn t follow national test algorithm. Our finding demonstrated that 50% testing points have no SOP for each testing points. In contrast to this findings study conducted in Ethiopia and South East Asia reported that from assessed laboratories 0% and 8% did not have SOPs respectively 1112.

    The average score of the testing points were 92.7%. Figure 1 This slightly lower than nine year survey conducted in Africa WHO region which indicated that 98.9% 13. This deference may be due to study participant difference and method used for HIV testing. Because our study participants included all testing points and they performed HIV testing using HIV rapid test but their study participants only laboratory professionals and they used HIV rapid test and ELISA to perform HIV rapid testing. Our findings demonstrated that 82.8% of testing points score 80 and above this lower than study conducted by Jean Louis and his colleagues which was 97.5% 14.

    Average performance of testing points in %

    Conclusion

    Conclusions and Recommendations

    High quality test results underpin accurate diagnosis and appropriate treatment for patients. But in the study area the score of PT result and coverage of training is slightly low comparing to other findings. Therefore following national testing algorithm to report client test results, training and monitoring are critical points to improve the PT score of testing points.

    Affiliations:
    Affiliations: