SEARCH

SEARCH BY CITATION

Keywords:

  • tuberculosis;
  • diagnostic delay;
  • gender;
  • socio-economic;
  • Africa

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

In-depth interviews regarding health seeking behaviour were conducted with 202 adults registered with pulmonary tuberculosis at the centralized Chest Clinic in Lusaka, Zambia. The median (mean) diagnostic delay was 8.6 (9) weeks, and was significantly associated with the following factors: female sex, lower education, more than six instances of health-seeking encounters, outpatient diagnosis of tuberculosis, and visiting a private doctor or traditional healer. More effective tuberculosis control interventions require novel methods of accessing women and less educated people. Decentralization of public tuberculosis care and improved integration with private sector health providers may also reduce diagnostic delay.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Tuberculosis control programmes are critical in managing the public health risk of this infectious disease. Minimization of the time from symptom onset to initiation of therapy, known as ‘diagnostic delay’, is important for successful tuberculosis programmes (Mori et al. 1992). Increased delay leads to more serious illness by the time a patient presents (Gibson et al. 1998) and to increased infectivity within the community (Lawn et al. 1997). The median diagnostic delay within urban centres in Canada (Wobeser et al. 1999), Malaysia (Liam & Tang 1997) and Ghana (Lawn et al. 1998) is 6, 12.5 and 17 weeks, respectively. Clearly improvement is needed, especially within resource-poor settings which have the greatest tuberculosis burden. Conflicting reports exist on how patient socio-economic characteristics, gender and the health services system affect diagnostic delay (Murray 1994). The purpose of this research is to clarify if these factors are associated with patients’ diagnostic delay for tuberculosis within a typical urban African setting.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

At the time of this study, public sector tuberculosis control services in Zambia’s capital city, Lusaka, were centralized at the Chest Clinic located at Zambia’s tertiary hospital. The Chest Clinic was a referral centre for the diagnosis and registration of tuberculosis inpatients and outpatients. Although services at the Chest Clinic were free, outpatients required referral from the hospital or community clinics which required payment of a fee or subscription to the national health insurance programme. Tuberculosis diagnosis was assisted through sputum microscopy and culture performed at the centralized government laboratory. Patients diagnosed at the Chest Clinic were simultaneously registered in the national TB programme, allowing them access to free TB drugs distributed through decentralized local clinics.

Over 9 weeks, adults registered with new pulmonary tuberculosis were referred to the study. Both smear-positive and -negative patients were included. When the number of patients awaiting interview was greater than three, the clerk temporarily stopped referral. This patient enrolment process did not embody any particular selection bias and minimized potentially important study-related delays for patients. In-depth interviews were performed with consenting patients focusing on socio-economic and care-seeking data, including the estimated time from symptom onset to tuberculosis diagnosis at the Chest Clinic. As outlined in the table, relevant variables affecting diagnostic delay were identified through literature review (Mori et al. 1992; Rubel & Garro 1992; Murray 1994; Hudelson 1996; Liam & Tang 1997; Liefooghe et al. 1997; Gibson et al. 1998; Lawn et al. 1998; Steen & Mazonde 1998). For each variable, two patients groups were defined (the median value was used to divide non-binary variables). The log-rank test was then used to compare the distribution of delays between each group (Harrington & Fleming 1982).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

A sample of 225 patients were referred to the study. Excluding 19 patients refusing participation (primarily because of personal time constraints) and four incomplete responses, complete data were obtained for 202 patients. These patients were typical young, productive, urban Zambians. The median (mean) diagnostic delay was 8.6 (9) weeks. This delay period includes the 3-week period of cough during which time tuberculosis would not be considered as a diagnosis, under current guidelines, unless haemoptysis was present. This is consistent with the measurement of delay used in other similar studies such as Lawn et al. (1998), Liam and Tang (1997) and Wobeser et al. (1999). As summarized in Table 1, significant socio-demographic factors associated with delay were: female sex and education of less than 9 years. Among health services factors, delay was associated with outpatient diagnosis of tuberculosis, having visited a private doctor or traditional healer, and more than six tuberculosis-related health-seeking encounters (a health-seeking encounter being a visit to a pharmacy, traditional healer or any formal health service provider). The difference in mean delay between each group was approximately 2 weeks for most of the significant factors, except for those patients with more than six health-seeking encounters and those who sought care from a traditional healer, for whom the difference in mean delay was approximately 3 and 6 weeks, respectively.

Table 1.   Factors affecting TB diagnostic delay in urban Zambia Thumbnail image of

Although tuberculosis is a great economic burden on patients (Gibson et al. 1998; Needham et al. 1998), economic factors were not significant using the indirect measures considered in this analysis. There was a trend, however, toward shorter delay among patients who could obtain external financial assistance.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Based on these results, recommendations for more effective tuberculosis control interventions may be considered within the setting of this study. Tuberculosis control programmes must take into account the special circumstances of women and people with less education to maximize these patients’ access to services. Longer delays for women correspond to more lost work days and lost income (Needham et al. 1998). Lower education among women may be a confounding factor in this univariate analysis. The heavy workload of women combined with their lack of mobility, independence and financial resources may be other potential explanations for this finding (Hudelson 1996). With high rates of TB–HIV co-infection in this population, the associated stigma may delay patients, especially women, in seeking care (Hudelson 1996). The barriers of transportation and time costs (Needham et al. 1998) could be reduced through proposed measures such as integration of tuberculosis screening services for women into existing primary health care or paediatric clinics already attended by women (Hudelson 1996). Such an intervention may also minimize the number of health encounters prior to diagnosis. Stigmatization, especially affecting women (Hudelson 1996), should be addressed through health education directed at women which reinforces the message that tuberculosis is curable.

Numerous patient health encounters are associated with diagnostic delay, as expected. Inpatient diagnosis of tuberculosis may reduce the number of health care encounters and consequently result in the shorter delay which was observed in comparison to outpatients. Within the public health services system, decentralization of tuberculosis diagnosis from the Chest Clinic to upgraded local health clinics may also reduce the number of health encounters which lead to delay (Lawn et al. 1998; Needham et al. 1998), and reduce waiting times because of the larger number of TB health providers. Decentralization is currently underway in Zambia through health sector reforms. However, difficulties with funding shortages, and supervision and training of personnel at many clinics have limited the ability to successfully decentralize services. These difficulties may offset the potential benefits of decentralization to tuberculosis control.

The delay associated with expensive private health sector consultation (traditional healers and private doctors) may be a result of both numerous unfruitful consultations and depletion of patients’ limited financial resources otherwise required for the tuberculosis diagnostic process (Gibson et al. 1998; Lawn et al. 1998; Needham et al. 1998). Education regarding the symptoms and signs of tuberculosis directed at these health providers may enhance case-finding and minimize diagnostic delay. Better integration of the private sector into the public sector tuberculosis control programme may also help. This could be accomplished through allowing private sector health care providers access to central laboratory sputum microscopy and culture services of the public sector tuberculosis control programme, and to patient registration for free TB treatment. Patients could then seamlessly pass between private and public sector health providers without redundant investigations, unnecessary paper work and the associated delays. Finally, other reasons that patients seek private sector health care, such as shorter waiting times, and more flexible and improved personal service (Waddington & Enyimayew 1990; Bichmann et al. 1991) need to be vigorously addressed within the public system. Expanded communication skills training for local public health service providers may also improve patient satisfaction with public-sector services.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Dale Needham acknowledges the financial assistance of the Farquharson Research Award (Medical Research Council, Canada), the Commonwealth Foundation and Lennox–Boyd Memorial Trust, and GlaxoWellcome. Susan Foster acknowledges the financial support of the UK Department of International Development (DfID) Work Programme on AIDS and Sexually Transmitted Diseases. Peter Godfrey–Faussett is a Beit Memorial Medical Fellow. The ZAMBART (Zambia AIDS-Related Tuberculosis) Project is funded by the Commission of The European Community, DfID and World Health Organization.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Bichmann W, Diesfeld H, Agboton Y, Gbaguida E, Simshauser U (1991) District health systems: users’ preferences for services in Benin. Health Policy and Planning 6 , 361370.
  • 2
    Gibson N, Boillot F, Jalloh H (1998) The cost of tuberculosis to patients in Sierra Leone’s war zone. International Journal of Tuberculosis and Lung Disease 2 , 726731.
  • 3
    Harrington DP & Fleming TR (1982) A class of rank test procedures for censored survival data. Biometrika 69 , 553566.
  • 4
    Hudelson P (1996) Gender difference in tuberculosis: the role of socio-economic and cultural factors. Tubercle and Lung Disease 77 , 391400.
  • 5
    Lawn SD, Shattock RJ, Griffin GE (1997) Delays in the diagnosis of tuberculosis: a great new cost. International Journal of Tuberculosis and Lung Disease 1 , 485486.
  • 6
    Lawn SD, Afful B, Acheampong JW (1998) Pulmonary tuberculosis: diagnostic delay in Ghanaian adults. International Journal of Tuberculosis and Lung Disease 2 , 635640.
  • 7
    Liam CK & Tang BG (1997) Delay in the diagnosis and treatment of pulmonary tuberculosis in patients attending a University teaching hospital. International Journal of Tuberculosis and Lung Disease 1 , 326332.
  • 8
    Liefooghe R, Baliddawa JB, Kipruto EM, Vermeire C, De Munynck AO (1997) From their own perspective. A Kenyan community’s perception of tuberculosis. Tropical Medicine & International Health 2 , 809821.
  • 9
    Mori T, Shimao T, Jin BW, Kim SJ (1992) Analysis of case-finding process of tuberculosis in Korea. Tubercle and Lung Disease 73 , 225231.
  • 10
    Murray CJL (1994) Issues in operational, social, and economic research on tuberculosis. In: Tuberculosis: Pathogenesis, Protection, and Control (ed. BR Bloom), American Society of Microbiology, Washington, DC, pp. 583–622.
  • 11
    Needham DM, Godfrey-Faussett P, Foster SD (1998) Barriers to tuberculosis control in urban Zambia: the economic impact and burden on patients prior to diagnosis. International Journal of Tuberculosis and Lung Disease 2 , 811817.
  • 12
    Rubel AJ & Garro LC (1992) Social and cultural factors in the successful control of tuberculosis. Public Health Reports 107 , 626636.
  • 13
    Steen TW & Mazonde GN (1998) Pulmonary tuberculosis in Kweneng District, Botswana: delays in diagnosis in 212 smear-positive patients. International Journal of Tuberculosis and Lung Disease 2 , 627634.
  • 14
    Waddington C & Enyimayew KA (1990) A price to pay. Part 2: the impact of user charges in the Volta region of Ghana. International Journal of Health Planning and Management 5 , 287312.
  • 15
    Wobeser W, Yuan L, Naus M (1999) Outcome of pulmonary tuberculosis treatment in the tertiary care setting – Toronto 1992/93. Canadian Medical Association Journal 160 , 789794.