- Research article
- Open Access
- Open Peer Review
This article has Open Peer Review reports available.
Characteristics and treatment outcomes of tuberculosis patients who “transfer-in” to health facilities in Harare City, Zimbabwe: a descriptive cross-sectional study
© Takarinda et al.; licensee BioMed Central Ltd. 2012
Received: 30 April 2012
Accepted: 12 November 2012
Published: 15 November 2012
Zimbabwe is among the 22 Tuberculosis (TB) high burden countries worldwide and runs a well-established, standardized recording and reporting system on case finding and treatment outcomes. During TB treatment, patients transfer-out and transfer-in to different health facilities, but there are few data from any national TB programmes about whether this process happens and if so to what extent. The aim of this study therefore was to describe the characteristics and outcomes of TB patients that transferred into Harare City health department clinics under the national TB programme. Specific objectives were to determine i) the proportion of a cohort of TB patients registered as transfer-in, ii) the characteristics and treatment outcomes of these transfer-in patients and iii) whether their treatment outcomes had been communicated back to their respective referral districts after completion of TB treatment.
Data were abstracted from patient files and district TB registers for all transfer-in TB patients registered from January to December 2010 within Harare City. Descriptive statistics were calculated.
Of the 7,742 registered TB patients in 2010, 263 (3.5%) had transferred-in: 148 (56%) were males and overall median age was 33 years (IQR, 26–40). Most transfer-in patients (74%) came during the intensive phase of TB treatment, and 58% were from rural health-facilities. Of 176 patients with complete data on the time period between transfer-in and transfer-out, only 85 (48%) arrived for registration in Harare from referral districts within 1 week of being transferred-out. Transfer-in patients had 69% treatment success, but in 21% treatment outcome status was not evaluated. Overall, 3/212 (1.4%) transfer-in TB patients had their TB treatment outcomes reported back to their referral districts.
There is need to devise better strategies of following up TB patients to their referral Directly Observed Treatment (DOT) centres from TB diagnosing centres to ensure that they arrive promptly and on time. Recording and reporting of information must improve and this can be done through training and supervision. Use of mobile phones and other technology to communicate TB treatment outcomes back to the referral districts would seem the obvious way to move forward on these issues.
Tuberculosis (TB) remains a disease of major public health concern the world over with most cases being found in the South-East Asia, African and Western Pacific regions . Emergence of the human immunodeficiency virus (HIV) epidemic has fuelled the TB epidemic worldwide, with one in ten TB cases in 2010 being estimated to be HIV-positive . The African region has been the worst affected by the HIV epidemic and accounts for over 80% of the global HIV-TB burden . Zimbabwe has not been spared as it is one of the 22 World Health Organisation (WHO) high TB burden countries  and currently has an HIV prevalence of 15.2% for the adult population .
The World Health Organization has for nearly 20 years implemented a Stop TB Strategy for global and national TB control, and one of its cornerstones is a standardised recording and reporting system. For smear-positive pulmonary TB (PTB) patients, at the end of a course of treatment there are 6 possible outcomes: cure, treatment completed, failed, died, lost to follow up and transferred out [3, 4]. A transfer-out is a patient who transfers from one reporting centre to another centre in a different reporting district and for whom the treatment outcome is unknown. This can be a sizeable problem: for the 2.5 million new smear-positive PTB patients registered in “DOTS” programmes globally in 2006, 3% or 75,000 patients were reported to be transferred out .
For the same year, the transfer-out rates in the African region were 4%, with individual countries reporting rates of between 0% – 16% . From the two most recent global reports in 2010 and 2011[1, 6], it is difficult to get a good estimate of the transfer-out problem as this treatment outcome category has been subsumed under a category called “not evaluated”: for the 2009 global cohort of treatment outcomes in new smear-positive PTB patients, the “not evaluated” proportion was 4% of 2.6 million patients registered , and this group probably contained a high number of patients who had transferred out.
Every patient who transfers out from the original registration unit should in theory “transfer-in” to a new TB registration unit in a different reporting district. However, the original registration unit maintains the responsibility for reporting on their treatment outcomes. Patients who are transferred into Harare City from another district carry with them a transfer out form from the referral district, which includes details such as patient name, referral hospital and district, sex, age, treatment category, type of TB and transfer-out date. These transfer-in patients are registered in the TB register at either of 2 infectious disease hospitals before referral for DOT at their nearest municipal clinic. Upon completion of TB treatment, the national guidelines specify that the outcomes of these transfer-in patients be routinely communicated back to the district from which the patient was transferred.
District TB coordinators for each district also follow up treatment outcomes of patients transferred out by issuing a treatment outcome request form to the transfer-in receiving health facility. If treatment outcomes of patients transferred out are unknown when cohort analysis is conducted, these patients have their treatment outcomes recorded as “transfer-out”. Including the treatment outcomes of patients transferred into a district as part of that district’s cohort is not allowed as this will result in more patients with treatment outcomes than patients notified and registered for that particular reporting period .
There are few data from national TB programmes about whether or to what extent this process happens. The aim of this study therefore was to describe the characteristics and outcomes of TB patients that transferred into Harare City health department clinics under the national TB programme. Specific objectives were to determine i) the proportion of a cohort of TB patients registered as transfer-in, ii) the characteristics and treatment outcomes of these transfer-in patients and iii) whether their treatment outcomes had been communicated back to their respective referral districts after completion of TB treatment.
This was a descriptive cross-sectional study design using registers and treatment cards.
This study was conducted in Harare, the capital city of Zimbabwe which has an estimated population of around 1.6 million . In Harare, National TB programme services are offered under the city health department, and hence the study included all 32 municipal clinics and 2 infectious disease hospitals which offer general health services integrated with TB treatment services . The municipal clinics are partitioned into either the southern or northern region, with each region consisting of an infectious disease hospital in which TB diagnostic services (smear microscopy and chest radiography) are performed for the surrounding clinics.
General diagnosis and management of TB patients
In Zimbabwe, TB is diagnosed according to national guidelines  which are based on the WHO TB treatment guidelines . Direct smear microscopy is the main method for diagnosing pulmonary TB, whereby suspected TB patients have their sputa collected at their nearest municipal clinic which are then sent for smear microscopy to the infectious disease hospital in their respective region. All confirmed TB patients are treated using standardised anti-TB regimens according to national  and international guidelines .
Monitoring is done clinically for smear-negative PTB and EPTB patients whilst those with new smear-positive PTB have sputum smears examined for acid-fast bacilli at 2, 5 and 6 months. Those with previously treated sputum smear-positive PTB have sputum examined at the end of 3,5 and 8 months. Smear-negative patients who complete treatment and smear-positive patients who complete treatment with or without negative smears are regarded as “successfully completing treatment”. Patients are offered HIV counselling and testing (opt-out provider-initiated) upon diagnosis of TB, and cotrimoxazole preventive therapy (CPT) is started together with anti-TB treatment for TB/HIV co-infected patients, provided there is no contra-indication. All HIV-positive TB patients are eligible for antiretroviral therapy (ART) initiation at Opportunistic Infections (OI) /ART initiating clinics at either of the 2 hospitals commencing patients on TB treatment in Harare.
Data were collected for all TB patients that transferred into Harare City between 1 January 2010 and 31 December 2010 from TB registers at the two infectious disease hospitals.
Data variables, data source and data collection
Patient data on those who transferred-in were abstracted from the TB registers between January 2012 and February 2012 using a data collection form, and variables that were collected included:- TB registration number, sex, age, type of TB, HIV status, TB treatment phase at time of transferring, date of transfer-out from the districts, date of transfer-in at the regional infectious disease hospital and the TB treatment outcome. Patient data in the TB registers were verified from individual patient files stored at their respective hospitals, and these patients were traced to their referral municipal clinics to establish if they arrived and were registered. District TB coordinators were also contacted by telephone to establish if treatment outcomes of their patients that had been transferred into Harare city had been communicated back to them. Individual TB registration numbers and respective dates of commencing TB treatment were used to trace these patients by district TB coordinators in their district TB registers.
Patient information on the data collection forms was coded and single-entered electronically into Epidata version 3.1 (The Epidata Association, Odense Denmark). The data were then exported to Stata version 10 (Stata Corporation, College Station, Texas) for data cleaning and statistical analysis. Medians and inter-quartile ranges were calculated for skewed continuous variables whilst proportions were generated for categorical variables. Time between date of transfer out from the district and arrival at a health facility in Harare was calculated by subtracting the transfer out date written on the transfer-out slip from the TB registration date recorded in the TB register of the receiving infectious disease hospital in Harare City. Comparisons between proportions was done using the chi-square test or alternatively the Fischers Exact test and Odds ratios (OR) with their 95% confidence intervals. Levels of significance were set at 0.05.
Ethics approval was granted locally by the Medical Research Council of Zimbabwe and the International Union Against Tuberculosis and Lung Disease (The Union). Confidentiality of information drawn from the patient records was ensured by excluding the patient names during data collection whilst all data collection forms were kept in a safe and secure place accessible only to the investigator.
Proportion of TB patients registered as transfer-in
There were 7,472 patients registered with TB, of whom 263 (3.5%) were recorded as transfer-in.
Characteristics and treatment outcomes of transfer-in patients
Demographic and clinical characteristics of transfer-in vs. Non transfer-in patients TB patient in Harare City – 2010
Patient Group (n(%))
Non Transfers-in (n=7029)
Age group in years
Category of TB
Type of new TB
PTB smears not done
Type of retreatment TB
Retreatment after default
Retreatment after failure
HIV test done
Of the 232 (95%) patients in whom an HIV test had been performed, 163 (71%) were HIV-positive. Of these, 143 (89%) were documented to be on CPT, but only 38 (23%) were documented as accessing ART during TB treatment. In comparison to non transfer-in patients, a greater proportion of transfer-in patients were HIV tested (95% vs. 63.9%, p<0.01), however a lesser proportion were diagnosed HIV-positive; 71% vs. 84%, p<0.01.
Characteristics of the Transfer process of transfer-in TB patients in Harare City, Zimbabwe
Time period between transfer-out and transfer-in to hospitals in Harare City (n=176)
< 8 days
8 – 31 days
Anti-TB treatment phase at time of transfer-in from the district (n=216)
Type of referral health facility from where patients were transferred-out (n=212)
rural mission hospital
urban municipal clinic
Referral site location (n=212)
Distance from Referral health facility to Harare City (km) (n=211)
Treatment outcomes of Transfer-in TB patients in relation to health facility and clinical characteristics
TB treatment outcomes n (%)
Sex (n = 263)
Age group (n = 260)
Type of TB (n = 256)
Time period between transferring from district and registration in Harare City (n = 176)
< 8 days
Type of referral health facility (n = 212)
rural mission hospital
Distance to referral health facility from Harare City (km) (n = 211)
41 – 150
151 – 300
Treatment phase at time of transfer from referral health facility (n = 263)
Treatment outcomes of Transfer-in TB patients in comparison to the non Transfer-in cohort
TB treatment outcome
Type of Patient n (%)
Non Transfers-in* (n=7010)
Communication of treatment outcomes of transfer-in patients back to referring districts
There were 225 of the 263 patients documented as being transferred in from 42 referring health facilities. Of these, there were only 3 (1.3%) patients who had treatment outcomes notified back at the referring health facility. For these 3 patients, the distance between the transfer-out and transfer-in facility was between 41 and 150 Km, the referring facilities were all rural and the 3 patients had successfully completed treatment.
This study, the first of its kind in Zimbabwe, shows that although a small proportion of patients transferred-in to Harare city out of the total number of patients registered, the management according to guidelines was poor with only 1.3% of treatment outcomes being notified back to the referring unit. Transfer-in patients generally had new pulmonary TB, had largely been HIV-tested and had a high HIV-prevalence rate, not very different from the non transfer-in patients in Harare city and in general among patients in Zimbabwe that have previously been reported on [8, 9].
Important lessons emerge from this study which can inform patient management. Most of the patients for whom there were data had transferred-in during the intensive phase of treatment, with 52% taking more than 1 week to transfer-out and transfer-in. For the 10% of patients in whom the transfer process takes longer, it is important that district TB officers ensure that the patients travel with an adequate supply of oral anti-TB drugs so that treatment is not interrupted. Although transfer-in retreatment patients were small in number, their management nevertheless is more complicated. The intensive phase of treatment necessitates intramuscular streptomycin for 2 months, and it would be generally prudent if these patients remained in their original treatment unit until the course of injections has been completed. Interruption of treatment is a potent cause of drug-resistance  and must be avoided at all costs.
Many patients transferred from long distances, often rural mission hospitals, and TB officers need to ensure that patients can afford their travel and are again well covered with the necessary drugs in case of travel delays or mishaps. Why such high numbers transfer-in from rural mission hospitals is not certain but may reflect many persons’ perception that church-related health services are of better quality than those run by government, and once patients feel better on anti-TB treatment they decide to return to their urban public health services for continuation of therapy .
Treatment outcomes of patients who transferred-in were rather similar to the non transfer-in cohort and to those reported from patients registered in their original units [8, 9], although there was a high proportion of patients not evaluated and with missing outcomes. The reasons for this are unclear, but may relate to poor documentation in registers and treatment cards or death, lost-to-follow-up and further transfer-outs which are not reported to the health facilities. Whatever the reasons, this is unsatisfactory and needs correction.
Finally, the majority of transfer-in TB patient did not have their outcomes communicated to their referral districts. This means the referring units would report these patients as “transfer-out”, while they could report on true outcomes if only communication had occurred. The conventional means of communication has been in the past through submission of treatment outcome request forms by the referral districts to the receiving districts through the postal system. Reasons for not communicating TB treatment outcomes are unclear. However, anecdotal reports have often attributed this to a shortage of TB treatment outcome request forms coupled with poor perceptions of postal services in Zimbabwe by health workers which are perceived as inconvenient, unreliable and an outdated mode of communication. There is therefore a need to move with the times and for the TB Control Programme to consider new cheap and feasible communication strategies such as use of a toll-free number for use by health-workers to their mobile phones .
There has been very little previous published work on the issue of transfer-in and transfer-out in TB patients. One study in Malawi showed that it was very common for TB patients to transfer-out, but the procedures for transferring–in and matching the two sets of patients was very poor . This is an area in need of improvement. The strengths of this study were that a large number of patients were evaluated and the work was done through the routine system.
Study limitations included the usual problem of completeness and accuracy of routinely recorded programme data and the fact that patients with unevaluated outcomes could not be traced to their respective physical addresses in order to establish their true outcomes. Comparison data on the non transfer-in patients were also obtained as aggregate data from national reports and could not be verified or authenticated.
In conclusion, there is need to devise better strategies of following up TB patients to their referral DOT centres from TB diagnosing centres to ensure that they arrive promptly and on time with drugs supplies uninterrupted. There is a need to improve on recording and reporting of information and this can be done through training and supervision. Use of mobile phones and other technology to communicate TB treatment outcomes back to the referral districts would seem the obvious way to move forward on this issue.
We would like to thank the Ministry of Health & Child Welfare and the AIDS & TB Unit, for their support and granting us authority to conduct the study. Funding for the study was obtained from the Expanded Support Programme for HIV/AIDS (ESP) whilst technical support was provided through the International Union Against Tuberculosis and Lung Disease (IUATLD). Kudakwashe Takarinda is supported as an operational research fellow from the Centre for Operational Research at the International Union Against Tuberculosis and Lung Disease, Paris, France.
- World Health Organisation (WHO): Global Tuberculosis Control: WHO report 2011 WHO/HTM/TB 2011.16. 2011, Geneva: WHOGoogle Scholar
- Zimbabwe National Statistics Agency (ZIMSTAT) ICF International: Zimbabwe Demographic and Health Survey 2010–11. 2012, Calverton: ZIMSTAT and ICF International IncGoogle Scholar
- Ministry of Health and Child Welfare: National Tuberculosis Guidelines. 2010, Harare: Ministry of Health and Child WelfareGoogle Scholar
- World Health Organization (WHO): Treatment of Tuberculosis Guidelines. 4th edition WHO/HTM/TB/2009.40. 2009, Geneva: WHOGoogle Scholar
- World Health Organization (WHO): Global tuberculosis control: epidemiology, strategy, financing: WHO report 2009 WHO/HTM/TB/2009.411. 2009, Geneva: WHOGoogle Scholar
- World Health Organization (WHO): Global tuberculosis Control 2010. WHO report 2010 WHO/HTM/TB/2010.7. 2010, Geneva: WHOGoogle Scholar
- Ministry of Health and Child Welfare: National Health Profile 2008. 2008, Harare: Ministry of Health and Child WelfareGoogle Scholar
- Takarinda KC, Harries AD, Srinath S, Mutasa-Apollo T, Sandy C, Mururungi O: Treatment outcomes of new adult tuberculosis patients in relation to HIV status in Zimbabwe. Public Health Action. 2011, 1 (Suppl 2): 34-39.View ArticlePubMedPubMed CentralGoogle Scholar
- Takarinda KC, Harries AD, Srinath S, Mutasa-Apollo T, Sandy C, Mururungi O: Treatment outcomes of adult patients with recurrent tuberculosis patients in relation to HIV status in Zimbabwe: a retrospective record review. BMC Public Health. 2012, 12: 124-10.1186/1471-2458-12-124.View ArticlePubMedPubMed CentralGoogle Scholar
- Mitchison DA: How drug resistance emerges as a result of poor compliance during short course chemotherapy for tuberculosis. Int J Tuberc Lung Dis. 1998, 2 (Suppl 1): 10-15.PubMedGoogle Scholar
- Green A, Shaw J, Dimmock F, Conn C: A shared mission? Changing relationships between government and church health services in Africa. Int J Health Plann Mgmt. 2002, 17: 333-353. 10.1002/hpm.685.View ArticleGoogle Scholar
- Harlow T: TB Net Tracking Network Provides Continuity of Care for Mobile TB Patients. Am J Public Health. 1999, 89 (Suppl 10): 1581-1582.View ArticlePubMedPubMed CentralGoogle Scholar
- Meijnen S, Weismuller MM, Claessens NJM, Kwanjana JH, Salaniponi FM, Harries AD: Outcome of patients with tuberculosis who transfer between reporting units. Int J Tuberc Lung Dis. 2002, 6 (Suppl 8): 666-671.PubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2458/12/981/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.