This study revealed that a substantial financial cost is incurred across pathways to TB care by patients and their families in a country with a significant burden of TB and poverty and where TB services are free of charge. The total median (IQR) household costs for TB care per patient including both direct and indirect costs amounted to USD129.2 (99.0-172.2)/ PKR14919. Overall, our study estimate is relatively comparable to a total median cost amounting to USD171 (75.6–387) ascertained in a population in China [17]. But our cost estimates were found to be comparatively lower total median costs than those reported in Ghana, Viet Nam and the Dominican Republic i.e. USD202, USD758 and USD742, respectively [18]. This could be possibly due to a heavy reliance on out of pocket expenses and inequality in healthcare distribution in low-income countries [19].
This study estimated indirect median costs amounting to USD52.0 (36.1–72.0)/ PKR5950 which is a significant contributor (42.10%) of the total household costs for TB care. This finding is comparable to a study conducted in a neighborhood which found estimated median indirect costs at USD78 [17]. A study conducted in Nigeria reported that indirect costs incurred per patient was USD79.13, which is relatively comparable to our study estimates [20]. Furthermore, looking at the pattern of costs contribution during different phases of TB care (Fig. 2), it is evident that indirect costs is relatively higher as compared to other costs categories during different phases and it is the highest contributor during treatment phase i.e. 62.7%. Our findings for higher indirect costs throughout is comparable to the findings in India which estimated an indirect costs by 54% of patients [21]. These costs could be due to the long course of disease and recovery pathway, sometimes occupied by complications leading to the loss of productivity. Among Thai adults, a study reported that approximately 20% experienced a decline in income due to the patients’ or their guardians’ reduced ability to work (23).
The substantially higher median (IQR) cost during the pre-diagnostic phase as identified in the study amounted to USD63.8 (44.7–90.5)/ PKR7377, which is 49% of the total household out of pocket payment for TB care. This could be due to informal care such as self-treatment and the practice of seeking health care from private facilities, which demand sizeable amounts of direct medical costs compared to public facilities where medical costs have been minimum and more tolerable. In our study, patients sought care from private hospitals (54.1%), dispensaries (5.8%), and drug stores (4.1%) when first contracting TB. Higher estimates before diagnosis was also reported in Ghana, Viet Nam, Dominican Republic and Zambia [18, 22]. Indirect median (IQR) costs in pre-diagnostic phase was USD29.2 (19.5–47)/ PKR3375 which has a significant contribution i.e. 50% out of total pre-diagnostic cost. Indirect costs have also been higher and contributed significantly as shown in earlier studies [18].
Diagnostic Median (IQR) costs were reported in this study at USD24 (15-35.5)/ PKR 2755 where direct medical costs shared 35.3% of it, which is significant. A systematic review for assessing the economic burden of TB reported direct medical median (IQR) costs for diagnostic phase amounting to USD30 [6]. This might be due to laboratory investigation costs and availing such services from private laboratories since people tended to have low confidence on the quality of services provided by public providers.
Besides the direct medical costs, indirect median (IQR) costs during the diagnosis phase reported in this study shared higher contribution USD 6.5 (4.0-9.5)/ PKR750. A comparable estimate of indirect median costs amounting to USD9.2 was reported in Ethiopia [23]. This might be because of time taken off from work for diagnostic tests and visits to receive the final diagnosis and for the prescription of treatment accordingly. During this phase, the patient either takes sick leave or resigns depending on the severity of illness, either way facing loss of income.
During the treatment and follow-up phase, the median (IQR) costs reported was USD10.5 (6.5–15.6)/ PKR1217 and USD15.7 (8.8–32.4)/ PKR1816, respectively. The finding is slightly higher as reported in Kenya where treatment costs amounted to USD3 only but comparable to follow-up costs amounting to USD23.43 as reported in Nigeria [24, 25]. It might be due to the greater share of indirect costs 62.70% and 39.8% during treatment and follow-up duration, respectively, as the patient had to take medications under the supervision or patients or their guardian had to visit the facility to pick up the drug. The follow-up costs in this study are slightly higher than the treatment costs since follow up visits require lab work to determine progress towards recovery.
Patients hospitalized at one point in time during the course of treatment were 17.6% (n = 91). This is less than the 33% and 23% reported in Ghana and Viet Nam [18]. Estimated hospitalization median (IQR) costs in this study amount USD 349.0 (147.2- 463.2)/ PKR40300. This estimate is considerably higher compared to findings reported in Ghana and Viet Nam which amounted to USD42 and USD118, respectively [18]. The possible reasons for the higher costs reported among hospitalized patients in our study might be due to the severity of symptoms which occurred resulting in direct hospitalization and causing high expenditures. It also could be associated with a relative delay in seeking care, i.e. after five to six weeks, and hence, a late diagnosis could lead to an advanced stage of the disease and turn into complications. Another reason could be the difference in costs of providing care from the provider’s perspective which might vary from one country to the other.
This study reported common coping modalities patients and their families have chosen to bear with the financial burden due to TB care. A substantial proportion of participants i.e. 94% used any mechanism available to finance their TB care thus further impoverishing their household’s ability to cope with the illness. Further, this study reported the distribution of different mechanisms of covering costs; the majority of the participants 59.5% utilized cash savings. However, our study finding was higher than one previously conducted study in Thailand which stated that 22% of participants were utilizing their own savings, this the most common mechanism found in Thailand [26].
Strengths & limitations: This study has certain strengths. This study is unique in Pakistan in reporting TB expenditure estimates in detail, comprising of costs incurred in different phases such as pre-diagnostic, diagnostic and treatment along with estimating indirect costs (loss of productivity). In addition, this study used a standardized questionnaire specifically developed for estimating patients’ costs for tuberculosis care comprehensively covering all aspects and phases for expenditure. This is among the few studies conducted on cost estimation for treatment of TB patient. The study sample was statistically calculated to determine the median costs from a regional study.
Certain limitations of the study need to be kept in mind while interpreting the costs. First, this facility-based study which was conducted in four public sectors where the majority of participants come from low and middle socio-economic groups. However, some patients who utilized the private sector and those unable to visit even the public sector hospital due to poverty are not included in this study. Second, this study was carried out in an urban setting where the socio-economic and socio-demographic status is different (may be higher) than in rural areas. Third, questions about costs and income are subject to recall bias and seasonal fluctuation, where validation of costs is difficult, particularly for those who had a number of visits to hospitals during pre-diagnosis and for those who required retreatment. Fourth, although patient cost surveys in different parts of the world somehow showed similar estimates, costs cannot be compared directly due to different methodologies employed in other studies and costs ascertained at different time periods.