Health System Challenges Affecting HIV And Tuberculosis Integration At .

Page 3 of 7both available but full integration is not possible.14Achievements in Malawi showed an increase in HIV testingof TB patients from 59% to 83%; furthermore, CPT and ARTprovision to HIV–TB co-infected patients improved from88% to 100% and 18% to 25%, respectively.13 Similarly, inKenya, there were higher percentages of HIV testing of TBpatients and CPT and ART provision than prior tointegration.13Integrated modelHuman immunodeficiency virus and TB services areprovided by the same trained healthcare provider at the samevisit, a ‘one-stop service’: TB clinic provides HIV treatment;HIV clinic provides TB treatment (Figure 1).13,14 This model isconsidered the most efficient and effective way to providecomprehensive TB–HIV services, and it is appropriate forsettings with high TB and HIV prevalence.13,14 Results inMalawi have shown an increase of 12% in HIV testing of TBpatients, from 85% to 97%, as well as an increase of 19% inART uptake, from 44% to 63%.13 Furthermore, there wasacceptability of the integrated services among patients.Similarly, in Kenya, integration of HIV–TB servicesdemonstrated improvements in HIV testing of TB patientsand provision of CPT and ART.13The purpose of this study was to conduct a rapid appraisal ofintegration of HIV–TB services in the provincial PHC clinicsin eThekwini District in 2015, with a view to determine theextent of the integration of HIV–TB services at facility andcommunity level and describe factors promoting andinhibiting integration of services.Original ResearchData collectionPrimary data on HIV and TB services at both facility andcommunity levels were obtained through key informantinterviews with the operational managers (OMs) andcommunity health worker (CHW) supervisors during clinicsite visits. A set of standardised data collection tools weredesigned by the researcher to ensure an appraisal of the type,extent and service delivery gaps of integrated HIV–TB servicesat PHC clinics in 2015. The data collection tools were critiquedand assessed for their validity by the NDoH HIV–TBprogramme and a panel of technical advisors. The first toolwas administered to OMs and the second tool to CHWs. Thetools consisted of structured questionnaires with predominantlyclosed questions and a few open-ended questions.Facility-level dataThe tool provided information on services offered; facilityinfrastructure; service organisation; policies and guidelines;clinical records; facility organisation; process of care; staffrotation and scheduling; human resources; equipment;medication and non-medical supplies; laboratory and othersupplies; monitoring and reporting; and communication,referrals and social mobilisation.Community health workersThe tool focused on service provision; training received;campaigns conducted; collaboration with non-governmentalorganisations (NGOs); and challenges experienced.A retrospective, observational, cross-sectional study designwas conducted.Secondary data was obtained by inspection of District HealthInformation System statistics, registers (TB, IPT, pre-ART andART) and the electronic databases TIER.net and ETR.net. Adata extraction form was used to collect data from the registers.This data included statistics on HIV-positive and TB patientsfor the period of the last quarter of the financial year 2014–2015, from 01 January 2015 to 31 March 2015. Clinical recordsof patients within the same period of time were analysed.Study settingData analysisThe study was conducted in the eThekwini MetropolitanHealth District in KwaZulu-Natal Province. The eThekwiniDistrict is densely populated (3 464 205), with the greatestconcentrations of the population settling in the South Region(41%). Primary healthcare services are offered at 110 clinics(43 provincial and 57 municipal clinics).Once the data was collected, the hard copies were stored in alocked file cabinet. The data on the hard copies weretransferred to a Microsoft Excel spreadsheet with the samecodes as the hard copies. Updated versions of the data didnot overwrite the older versions but were renamed usingcodes to ensure easy data tracking. The Excel spreadsheetwas only accessible to the research team. Double entry of thedata was done by the researcher and the research assistantsto ensure correct data entry and to maintain the integrity ofthe data. Descriptive statistics were computed, using theSTATA software system, version 13.0. Descriptive analysis ofvariables was expressed as frequencies and percentages forcategorical variables.Research methods and designStudy designStudy population, sample and sizeThe target population comprised all provincial PHC clinics ineThekwini Health District. Ten facilities (three best, threeintermediate and four poor performing facilities) werepurposively sampled by the PHC supervisors to accommodateeach subdistrict as well as to contain facilities that achievedor exceeded the TB performance indicators, in addition tofacilities that achieved some of the targets and others that didnot achieve any of the targets in 2013–2014.15http://www.phcfm.orgAnalysis of the open-ended questionsThe principal investigator analysed the open-ended questionsthemes, and emerging themes were identified and coded.Open Access

Page 4 of 7The results were reported as a percentage of facilities thatresponded with a certain thematic code.Ethical considerationsInformed consent was obtained from the participants. Ethicalapproval to conduct this study was obtained from theBiomedical Research Ethics Committee of the University ofKwaZulu-Natal (KZN) (reference number BE257/16).Permission to conduct the study was obtained from theeThekwini District Health Office and the KZN ProvincialDepartment of Health research unit.ResultsOriginal Research(NIMART) at six clinics. Similarly, fewer than 50% of nurseswere trained regarding the national TB guidelines. Themajority of clinics trained fewer than 25% of HCWs regardingTB infection control (Table 2).Seventy per cent of clinics rotate staff across different servicesevery 6 months.Challenges for staff rotation across different services wereattributed to deficiency in skills (80%); staff shortage (50%);and high turnover of staff, absenteeism and personalpreferences (30%).Medicine supply and equipmentHuman immunodeficiencyvirus–tuberculosis collaborative activitiesFacility levelOnly two clinics complied with the mandated integratedmodel for HIV–TB services. Three clinics were partiallyintegrated. The remaining five clinics were still implementinga stand-alone model (Table 1).Eighty per cent of clinics experienced a stock-out of ARTdrugs during the review period, while 30% of clinics reporteda TB medication stock-out. In relation to TB care, all clinicshighlighted stock-outs of sputum containers and N95 masks,compromising TB screening and infection control. RegardingHIV care, stock-outs of HIV testing kits and ART impedeearly HIV testing and early initiation of ART.Patient records and information systemCommunity levelThe overall integration of HIV–TB services at communitylevel is suboptimal. In the financial year 2014–2015, sevenclinics conducted awareness campaigns; however, these werefocused on vertical programmes. All the clinics haveestablished linkages with community traditional healers topromote HIV–TB integration. One clinic (1/9) establishedcommunity support groups and community adherence clubsfocused on ART. Three clinics have support groups focusedon vertical programmes. None of the clinics provide CPT andIPT at a community level. Challenges reported by the CHWsincluded staff shortage, high workload and lack of personalprotection equipment against TB infection.Factors influencing human immunodeficiencyvirus–tuberculosis integrationHealthcare workforceAvailability: The ratio of professional nurses (PNs) per100 000 outpatient department (OPD) population wasconsistent across four clinics ranging between 10 and 20;however the ratio of PNs per 100 000 catchment populationvaried across clinics, with the majority of clinics with ratiosabove 30 PNs per 100 000 catchment population.Training: Capacity building for HIV–TB integration wasinadequate at the selected PHC clinics. Fewer than 50% ofPNs were trained in nurse-initiated management of ARTTABLE 1: Frequency table depicting integration model in the selected primaryhealthcare clinics in eThekwini Health District, Durban, in 2015.Integration modelGovernance and leadershipOnly three clinics had a copy of the NSP. The majority ofclinics briefed the clinic staff, support partners and cliniccommittee on HIV–TB integration (see Table 3).Tuberculosis infection controlThe TB infection control practices were well implemented inthe clinics, with over 80% of clinics complying with eachactivity. Only one activity, triage, had fewer than 90% clinicscompliant (see Table 4).InfrastructureThe infrastructure in 50% of clinics in the current studycomprised seven or more consultation rooms, with threeTABLE 2: Percentage of trained healthcare workers at selected primaryhealthcare clinics in eThekwini Health District, Durban, 2015.Variable†Clinics T111556111222110%2014 TB Management778111001111106860000114330ClinicsnAll clinics are equipped with registers pertaining to TB, TBidentification, and HIV counselling and testing. Ninety percent of the clinics are equipped with pre-ART and IPTregisters. However, ART registers are available in 80% ofclinics. Only 20% of the clinics are using one fully integratedfile per patient. TIER.net was operational at nine clinics;however, 40% of clinics were not using ETR.net as yet.Stand-alone550IPC 2007Partially integrated330Fully integrated220HCWs, healthcare workers; NIMART, nurse-initiated management of ART; TB, tuberculosis;IPC, infection prevention and control.†, Percentage of HCWs trained regarding the following guidelines.http://www.phcfm.orgOpen Access

Page 5 of 7TABLE 3: Frequency table relating activities implemented to engage relevantstakeholders at selected primary healthcare clinics in eThekwini Health District,Durban, 2015.ActivitiesClinicsN%Briefing of staff on HIV/TB integration10100HIV–TB integration team990Briefing of clinic committee on HIV–TB integration880Briefing of NGOs and CBOs on HIV–TB integration10100Staff (monthly)10100With NGOs, CBO, HBC (monthly)550HIV–TB integration meetingsHIV–TB, human immunodeficiency virus–tuberculosis; NGOs, non-governmentalorganisations; CBOs, community-based organisations; HBC, home-based carers.TABLE 4: Frequency table demonstrating the tuberculosis infection controlpractices at selected primary healthcare clinics in eThekwini Health District,Durban, 2015.IPC activityTriageClinicsN%880Cough hygiene990Respiratory hygiene10100Ventilated sputum collection area10100Open windows10100Fans990IPC, infection prevention and control.clinics having ten or more rooms. Furthermore, the numberof consultation rooms in the two integrated clinics was 6 and10. Infrastructure challenges were noted to be barriers.Twenty per cent of clinics attributed the lack of integration offiles to infrastructural challenges.DiscussionSouth Africa adopted the full integrated model for HIV–TBservices. In the current study, only 20% of clinics hadestablished a fully integrated service model, while 50%clinics were still implementing a stand-alone model. Thesuccessful integration of HIV–TB services is dependent oninfrastructure, staff numbers and capacity, TB infectioncontrol, drug provision and stakeholder engagement.13InfrastructurePrevious studies indicate that three to four consultationrooms are required for integrated HIV–TB care.13 Althoughthe clinics had four or more consultation rooms, 90% of theclinics considered the number of consultation roomsinsufficient to provide HIV–TB integrated services.Infrastructure appeared to be a key constraint for integration;a similar finding in the 2014 joint review discussedinfrastructural challenges compromising effective TBinfection control.12 However, a possible explanation points todifficulties reorganising the location of the services toenhance integration as the historical mechanism of providingTB–HIV services as vertical services located in differentplaces at facility level is an obstacle for the provision ofintegrated TB–HIV services.16 Infrastructural challenges mayinhibit the implementation of a fully integrated model forHIV–TB services and maintain the current challenges of losshttp://www.phcfm.orgOriginal Researchto follow-up, long queues, poor adherence to treatment bypatients and patients’ dissatisfaction with the currentfragmented service delivery.16Human resourcesAvailability of staffThe staffing ratio of PNs per 100 000 OPD population weresimilar among the clinics. In contrast, a CHW shortage waspresent at all clinics, compromising adequate coverage ofhouseholds. Therefore, staff numbers were only assessed as aconstraint for integration at community level. This constraintwas also highlighted in a 2012 South African study thatinterviewed 29 health managers and NGOs, which revealedhuman resource capacity as a major barrier at communitylevel, reflecting the increased volume of HIV-associated TBcases, funding restraints hindering the creation of posts anddifficulties recruiting and retaining staff.16 Furthermore, theclinics experienced difficulties rotating staff regularly becauseof staff preferences, work teams and shifts, skills shortageand absenteeism. The long periods between staff rotation donot allow staff to practice the skills gained from in-service orformal training, thus hindering task-sharing and efficientintegration of HIV–TB services.Capacity building: Capacity building for integrated HIV–TBcare was inadequate for NIMART, TB infection and TBmanagement, potentially preventing task-shifting fromdoctors to nurses as mandated by national guidelines10,16 anddown-referral to community level because of lack ofstandardised training. These constraints of limited taskshifting and -sharing are mentioned in other South Africanstudies, such as the 2015 qualitative study on clinicians’perceptions and patients’ experiences of ART integration inPHC clinics, which stated that training was inadequate forPHC skills, national ART guidelines and NIMART, and a2012 qualitative study on health system barriers toimplementation of collaborative TB and HIV activities thatreferred to inadequate NIMART training.16,17Capacity building and availability of staff are likelyconstraints for integration at both facility and communitylevels. The low number of trained healthcare providerscompounded by the high volume of patients experienced bymost clinics could lead to poor quality of care, high turnoverof staff and burnout, and significant staff absenteeism, hencecounteracting the potential benefits of fully integrated HIV–TB care. These findings are echoed in the 2014 joint review12and a qualitative study conducted in Sisonke District inKwaZulu-Natal.16Tuberculosis infection controlThe fear of TB infection spread could be a possible constraint.Forty per cent of clinics stated that triaging was conductedbut not consistently, a constraint in keeping with a 2013 SouthAfrican study across 127 PHC clinics in three districts, whichfound that 48.8% (N 127 PHC clinics) did not separatecoughing patients from other patients.18 All the clinicsOpen Access

Page 6 of 7indicated stock-outs of N95 masks, a challenge consistentwith findings from the 2014 joint review12 and the previoushighlighted 2013 South African study in three districts.18Moreover, the lack of N95 masks affected collaborativeactivities at community level. The risk of TB spread is furtherexacerbated with the inadequate IPC training of HCWs,consistent with a 2013 South African study that revealed thatonly 33% PNs and 7.2% of CHWs had received training onTB infection control.18 Poor IPC will compromise nationalefforts to control the TB epidemic. Furthermore, PLWHIV areat an increased risk of TB infection acquisition and mortalitycompared to the general population. Therefore, fullyintegrated HIV–TB services necessitate effective TB infectioncontrol to prevent nosocomial transmission of TB ashighlighted in the Extensively drug-resistant TB (XDR-TB)outbreak in Tugela Ferry in KwaZulu-Natal in 2006.19Medicine supplyMedicine supply challenges included stock-outs of ART andTB treatment as indicated by 80% and 30% of clinics,respectively, potentially inhibiting integration of HIV–TBmanagement as highlighted in a 2011 South African case studyon obstacles and solutions regarding HIV–TB integration.19Governance and leadershipAll the clinics engaged their staff, clinic committee andpartners in the process of integration of HIV–TB services. Inspite of this level of stakeholder engagement, half of the clinicsdid not complete the process of integration. Stakeholderengagement at community level was mainly focused onvertical programmes, a constraint discussed in the 2014 jointreview, where many facilities missed opportunities to integrateTB and HIV community-based activities.12 In addition, therewas inadequate establishment of community support groups,in contrast to the 2014 Joint Review, which stated that therewas good practice of support and capacity building to localcommunity organisations through NGOs in some of theprovinces and facilities visited.12 The lack of advocacy,communication and social mobilisation for integrated HIV–TBservices at community level will inhibit creating a demand forintegrated HIV–TB services at all levels of care, thus impedingthe integration process.6Implications of study findingsA systematic review conducted on observational studiesusing secondary data has found several benefits fromintegration of TB–HIV integration in sub-Saharan Africa.20Implementing integration at facility and community levelrequires a combination of clinical integration (TB and HIVcare, treatment, diagnostic testing and health educationactivities taking place concurrently) and organisationalintegration (facility-level resources [e.g. staff, infrastructure,space, patient files and data systems] and processes [e.g.patient flow] are integrated).21 As this study has highlighted,full integration is being hampered by organisational factors.However, these factors are not insurmountable, as many ofhttp://www.phcfm.orgOriginal Researchthe highlighted challenges require minor modification inprocesses at facility level. Although the South Africanguidelines explain what is needed to integrate TB and HIVservices, the ‘how’ aspect is lacking. A quality improvementmodel using facility-based multidisciplinary teams togetherwith change management at facility and community level isrequired to ensure implementation and sustainability ofHIV–TB integration.Study limitationsAlthough due diligence was exercised in maintaining thescientific integrity of this appraisal, it is limited by the qualityof data, the selection bias because of non-probability samplingand most importantly the small sample size. These limitationscompromise the generalisability of these findings to othersettings and contexts.The quality of data was a limitation to the study because ofseparate registers and notes for ART and TB care requiringthe same information. Frequent data recording results inincomplete data recording in registers. In addition, althoughTIER.net (for ART clients) is implemented at all clinics, ETR.net (for registered TB clients) is still not implemented at 40%of clinics.Conclusion and recommendationsFour factors were identified as possible constraints forintegration of HIV–TB services, namely reorganisation oflocation of services, medicine supply, TB infection controland inadequate capacity building. The factors identified aspossible promoters included structural infrastructure,staffing ratios and stakeholder engagement.Human immunodeficiency virus and TB integration willneed to be improved by addressing the systemic challengesaffecting health service delivery, including strengtheningsupervision, training, and medicine supply and adopting aquality improvement model.AcknowledgementsThe authors wish to acknowledge the district manager,primary healthcare (PHC) supervisors and PHC facilitymanagers for their support in c

the need to visit two clinics. Nevertheless, it serves as a first step in the process of integration.13 Partially integrated model Some HIV services are provided in clinics and some TB services are provided in HIV clinics, but co-infected patients must still visit two different clinics served by different staff to