Determination of Direct and Indirect Costs Incurred by Tuberculosis Patients During Diagnosis and Treatment in Urban Areas of South Gujarat: A Mixed Method Approach

Abstract

Though many initiatives and monetary benefits are incorporated under RNTCP/NTEP, many patients might incur some out-of-pocket expenditure (OOP) related to diagnosis, treatment, and hospitalization. Such costs lead to further poverty and default. This study estimated OOP costs. A cross-sectional mixed method study was conducted in 2020. Data were collected from two selected UHCs (both public and private sectors) from all eight administrative zones. A total of 278 newly registered drug-sensitive tuberculosis patients at different stages of treatment were enrolled, and 18 IDIs were done after obtaining the consent. Among 278, 231 (83%) were seeking the treatment from the public sector and 47 (17%) from the private sector. The average direct, indirect, and total costs were Rs. 8812, Rs. 4825, and Rs. 13,637, respectively. Extra food and supplements are the major field of expenditure for those enrolled in the public sector. Higher costs were incurred by the private sector patients. Longer distances, a long waiting time, belief systems, and unavailability of facilities or drugs were the common reasons for not visiting the public sector. IDI results also supported the cost heads. The majority of the expenses occurred at the private settings before diagnosis. IDIs suggested to changes in the programmatic approach toward migrants, industrial workers, and women.

INTRODUCTION

Tuberculosis (TB) is an infectious disease that disproportionately affects the poor. TB programs therefore need to ensure that the economically and socially disadvantaged groups face fewer barriers to improve seeking treatment. Some monetary benefits are provided under RNTCP to meet with the expenditure related to the treatment and nutritional needs. Still, TB programs need to ensure that TB does not stand at the beginning of a spiral into poverty. Out-of-pocket (OOP) costs for public and private health-care services may stand at the starting of a spiral which leads to poverty for many families and exacerbate the poverty of the already poor. This situation has been termed the “the medical poverty trap”.[1]

India is a country with a high burden of TB[2] with challenges related to treatment coverage.[3] Despite TB diagnosis and treatment services being free under the national TB programs, patients incur significant direct medical, direct nonmedical, and indirect costs due to TB care.[4,5] Under RNTCP in India, the program expects quality improvements and efficiency benefits contributing to significant cost savings. By taking a detect–treat–prevent–build approach, the national program can achieve significant positive change and make a real difference in the lives of the many people it serves.[6] The Government has also taken some initiatives like implementation of ‘Nikshay Poshan Yojana’[7] and incentives to private practitioners[8] for engagement in the cascade of care.

Though many initiatives and monetary benefits are incorporated under RNTCP, many patients might incur some OOP expenditure while visiting health facilities related to diagnosis, treatment, and hospitalization. The OOP cost can be incurred in forms of 1) charges for health services; 2) transport, accommodation, and subsistence; and 3) lost income, productivity, and time. Urban areas have some challenges like migration, industrial workers, and working women. These groups are more vulnerable in the context of getting diagnosed and treated.[9,10,11,12,13,14] Thus, being medically and socially vulnerable, these groups require specific attention in the context of their challenges and coping strategies related to treatment of TB.

Knowledge of financial burden on the beneficiary might help in reducing and removing OOP expenditure and thus might increase adherence and treatment completion. This study was planned to estimate the OOP costs incurred by the patients suffering from TB and to explore the areas for better programmatic management and improvement of overall public health in India.

Objectives

• To estimate the OOP cost incurred by the patients suffering from TB

• To explore the reasons for OOP expenditure and response by the study participants for expenditure met (qualitative)

METHODS

A cross-sectional study was conducted during January to March 2020 using a pretested semistructured questionnaire. Data were collected, including the patients (n = 278) who were registered at the City Tuberculosis Office, Surat. Personal interviews were conducted by trained personnel at residences of participants who were registered at two selected UHCs (both public and private sector patients) from each zone out of eight administrative zones. Selection of UHC from each zone was done randomly.

Patients were selected by the ‘Purposive’ method of Sampling. For enrolment in this study, participants at the following different stages of treatment were selected from each UHC: (1) At diagnosis, (2) on completion of the intensive phase, (3) on completion of 4 months of treatment, and (4) treatment completed, so that the OOP cost at these different stages can be documented among these selected public sectors and private sectors; cases of pulmonary and extrapulmonary TB were included. The proportion of patients enrolled in all the categories was near to one fourth. Thus, the cost obtained represents an average scenario. Thus, selections of samples were as follows: 231 from the public sector and 47 from the private sector, who were newly registered patients with drug-sensitive TB (n = 278).

Participants were explained about the purpose of the study and consent process. They were also informed about benefits and harms related to their participation in this study. After obtaining written consent, willing participants were enrolled for this study and interviewed by trained personnel.

A qualitative approach was adapted to understand the various expenses by different groups residing in the urban area. Costs, Challenges, and coping mechanisms of city-specific vulnerable groups were addressed though IDIs during October 2020 to February 2021. The selected groups were women having a family and children, factory workers, and migrants.

Exclusion criteria

The following participants have been excluded from this study:

• Patients aging less than 18 years

• Pediatric, human immunodeficiency virus-positive, and diabetic comorbidity

• Patients having past history of TB

Data entry and analysis

• Quantitative survey: Data were entered in MS Excel and analyzed with the help of SPSS. Descriptive statistics and cost analysis for estimation of direct, indirect, and total costs was done.

• Field notes and/or audio recordings were utilized to prepare transcripts. An appropriate qualitative approach was adapted (IDI).

• Costs in Indian Rupees were converted to US Dollars by dividing them by 73 (March 2021) as per the prevalent exchange rate for the currency for comparison at the international standard.

Ethical consideration

Permission from the Institutional Ethics Committee (IEC) of Government Medical College, Surat, was obtained before proceeding with recruitment of study participants. Written consent was obtained after explaining the patients about the purpose and procedure of this research. The audio recorded or written consent was taken from the participants of IDIs. Confidentiality was maintained while data collection, data entry, and analysis; the data without individual identifiers were shared with the coinvestigators for analysis purposes.

RESULTS

A total of 278 new cases of TB patients were included in this study. Among them, 231 (83%) were seeking the treatment from the public sector and 47 (17%) from the private sector. The majority, 239 (86%), were suffering from pulmonary illness, while the rest, 39 (14%), from extrapulmonary illness. Among the total 278 patients, 149 (53.6%) were males and 129 (46.4%) were females. The age of the study population ranged from 18 years to 78 years. The mean age was 32.2 years, and the standard deviation (SD) was 14 years. The majority, 174 (64%), were married, and the rest, 98 (36%), were unmarried. The average per-capita monthly income was less than 5000 among the majority (67.7%) of participants; 33.1% were earning Rs. 5001–10,000, 3.6% were earning Rs. 10,001–15,000, 1.4% were earning Rs. 15,0001–20,000, and 3.2% were earning more than Rs. 20,000 [Table 1].

Table 1:

Sociodemographic characteristics of the patients in this study

No.	Characteristic	n	Percentage
Sex (n=278)
1	Male	149	53.6
2	Female	129	46.4
Marital status (n n =278)
1	Married	174	64
2	Unmarried	98	36
Age Mean – 32.2 years, SD – 14 years, Range 18–78
Educational status (n=278)
1	Not studied at all	45	16.2
2	Primary (1 to 8)	119	42.8
3	Secondary (9-10)	65	23.4
4	Higher secondary (11-12)	25	9
5	Graduation	19	6.8
6	Postgraduation	5	1.8
Occupation (n=278)
1	Job - Skilled	78	32.5
2	Job – Nonskilled	32	13.3
3	Business	14	5.8
4	Housewife	69	28.8
5	Other	47	19.6
6	Missing data	38
Type of family (n=266)
1	Nuclear	107	40.2
2	Joint	136	51.1
3	Three generation	2	0.7
4	Household	21	7.9
Native place (n=274)
1	Within Gujarat	208	75.9
2	Outside Gujarat	66	24.1

Average cost incurred by patients regarding different heads of expenditure

Patients were asked about their expenditure incurred with different stages and reasons related to diagnosis and treatment of TB. Average costs were found as follows: the pretreatment cost was Rs. 5421, the treatment cost was Rs. 1920, the cost for nutritional supplements like food rich in proteins and vitamins was 1079, and the cost due to other coexisting illnesses was 155. Thus, the average direct cost was Rs. 8812. The average of different indirect costs was as follows: a wage loss of Rs. 4216, a supporter-related cost of 298, and a total indirect cost of Rs. 4825. The average OOP expenditure was computed, being Rs. 13,637 [Table 2].

Table 2:

Average cost incurred by patients regarding different heads of expenditure

No	Head of Expenditure	Mean (Rs)	SD (Rs)	Mean (USD)	SD (USD)
Direct cost
1	Pretreatment consultation	1566	5741	21.5	78.6
2	Pretreatment investigations	8063	22377	110.5	306.5
3	Pretreatment hospitalization	1879	8978	25.7	123.0
4	Pretreatment transport	34.5	40	0.5	0.5
	Pretreatment cost	5421	17955	74.3	246
1	Hospitalization during treatment	883	5309	12.1	72.7
2	Transport during treatment	49	113	0.7	1.5
3	Consultation during treatment	204	1634.7	2.8	22.4
4	Medicine during Treatment	447	3461	6.1	47.4
	Treatment cost	1920	8890	26.3	121.8
1	Nutritional supplements cost	1079	1869	14.8	25.6
2	Other illness	155	845	2.1	11.6
	Total direct cost	8812	24204	120.7	331.6
Indirect cost
1	Wage loss	4216	11697	57.8	160.2
2	Supporter-related	298	1058	4.1	14.5
	Total indirect cost	4825	14877	66.1	203.8
OOP cost
	Total OOP cost	13637	30008	186.8	411.1

*1 USD = Rs. 73 as per March 2021

Here, a high SD was obtained. It suggests disparity of expenditure among the patients. As most of the patients had to spend less than 5000, for those who needed to spend more, the amount shouted, which is reflected on computation of SD as the data were collected during January to March 2020 before initiation of lockdown.

Sectorwise distribution of total OOP cost among the patients

On cross-tabulating the total cost and sector of seeking treatment, it was observed that those in the private sector incur more OOP expenditure as compared to public sector patients. This difference was found to be statistically significant [Table 3].

Table 3:

Sectorwise distribution of total OOP cost among the patients

No.	Sector total cost	Public	%	Private	%
1	<5000	160	58	5	2
2	5001-10000	20	7	13	5
3	10001-15000	13	5	8	3
4	15001-20000	7	3	2	2
5	>20000	31	11	19	7

Chi-square value – 0.000 (P<0.001)

No statistically significant difference was observed between OOP expenditure and sex, educational status, occupation, per capita income, native, and type of the illness.

A 49 years old patient of extrapulmonary TB who studied up to secondary school, earning Rs. 6000 monthly, reported selling a vehicle costing Rs. 18,000. Another 49 years old female patient having a family income of Rs. 50,000 reported that she sold ornament costing Rs. 50,000 for coping with the expenditure due to illness. Both were being treated by the private facilities. One of the patients reported that he chose biopsy getting done with the private sector when it was suggested by the public sector (IDI participant).

Reasons for not visiting Government health facilities

On asking the patients the reasons for not visiting Government facilities, the most common reason was longer distances, reported by 29.6%, followed by mistrust on Government health facilities by 14.5%, a long waiting time by 11.9%, belief systems by 10.7%, unavailability of facilities or drugs by 6.3% each, and other reasons by 20.7%. On inquiring about other reasons for not visiting Government health facilities, preference of a family doctor was reported as the most common reason, while a few wanted to visit the nearest facility only; a few followed relatives’ or friends’ suggestions; a few perceived not well-managed facilities; one patient went to a private clinic because he himself was working there. “More time is required if long queues in government.” (IDI participant) [Figure 1].

Figure 1.

Reasons for not visiting Government facilities (N = 159)

Patients who visited private facilities before or during the treatment incurred higher costs, which was statistically significant. This was suggestive of lower OOP among those enrolled in the government setup; but longer distances, mistrust, long waiting times, belief systems, and unavailability of facilities or medicines with them might have prevented them from approaching Government sector healthcare facilities.

Highlights from IDI

Migrant patients, women, and industrial workers incurred high costs related to diagnosis, investigations, and treatment in the private setup before they were enrolled in the public sector. Other fields of expenditure were hospitalization in private settings, consultation of private practitioners, buying medicines, and so on. Within the public sector, patients had to spend for transportation, food and other supplements, costs related to supporters (spouse, daughter, friend) like transportation, food, and so on [Table 4]. Borrowing money from family members and friends was one the coping mechanisms to the OOP expenditure for migrants and Industrial workers. One of the female participants did not face coping issues because family members supported. “Everybody in my family supports me in taking the treatment. The financial burden is increased on my family due to my illness. They help me getting good food from outside.” (comment from one woman). A few industrial workers reported getting medicines promptly on reaching the facility, whereas a few reported long queues and waiting times. One participant responded, ‘I did not get any salary for about one and a half month as I could not go for my job. They deduct Rs 400 of daily wage when I don’t go for job, sometimes also when I need to go to the hospital for consultation or treatment.’ (comment from one Diamond worker).

Table 4:

Scenario of IDI participants of different categories

No	Theme	Migrants Average cost (Rs)	Females with Children and Family Average cost (Rs)	Industrial workers Average cost (Rs)
1	Diagnosis and treatment in private hospital	100000	2000	17000
2	Hospitalization in private	50000
3	Investigations in private setup	2525	575	500
4	Private practitioner consultation	2477	650	1933
5	Hospitalization in government	700
6	Transportation to government facility	63	50	35
7	Food per day	157	83	350
8	Supporter	800		1000
9	Medicines	250	1500
10	Financial help	35000		9500
11	Job loss	2500		19000
12	Time of waiting at government facility	45 min

*Blank cells suggest that the group of the participants did not answer related to the field of cost

DISCUSSION

The average OOP expenditure as recorded in this study in INR and USD, respectively, is as follows: direct cost 8812 and 120.7; indirect cost 4825 and 66.1 and total cost 13,637 and 186.8. In a study conducted in Karnataka by Poornima M P et al.,[15] the average prediagnostic cost was recorded as Rs. 3800, the direct medical cost was Rs. 5000, the direct nonmedical cost was Rs. 3000, and the indirect cost was Rs. 300. The current study presented a similar scenario related to the direct cost, but higher indirect costs were reported. Different study settings might be the probable reason. Sinha P et al.,[16] in one of their research, reported direct and indirect costs of 46.8 and 666.6 USD, respectively. Higher direct and lower indirect costs were reported by this study. The study by Sinha P et al. was conducted in rural settings, whereas in this study, study settings is an urban area might be one of the reasons of higher OOP expenditure among the patients enrolled in this study. In a study in Tamil Nadu, Veesa K S et al.[17] reported patients having incurred an average prediagnostic cost of 38.78 USD with delays at the various points of the healthcare delivery system. Prasanna T et al.,[18] in one of the research studies, reported direct and indirect costs of 65.3 and 50.2 USD, respectively. The current study recorded higher costs and probable reasons for being conducted in an urban area.

Though the system is well organized through UHCs, it is also noteworthy that patients reported long distances as a reason for not visiting government facilities. This suggests requirement of stronger IEC about availability of facilities in their vicinity.

CONCLUSION

Though free services for investigations, diagnosis, treatment, and care are available through the public sector, patients in the current study reported having incurred high OOP expenditure. The majority of the expenses occurred at the private settings. This is suggestive of the factors like longer distance, unawareness about services offered by national health programs, mistrust on public sectors, trust on family doctors, long waiting time, belief system, and unavailability of facilities or medicines might have prevented them from approaching public sector healthcare facilities. Extra food and supplements were the major field of expenditure for those enrolled in the public sector.

IDIs suggest lower spending (the majority were housewives) with the private sector for women as compared to men; this might be an area for availing the services offered to the women and creating supportive systems. As a migrant group lacks social support in the city, it might be the reason of higher OOP expenditure and special attention can be given to the vulnerable group.

Healthy workers are the backbone of healthy industries. TB being an air borne infection, for people working in different industries, their employer may take some initiatives so that workers can be rapidly registered directly with the public sector. Employee welfare in relation to TB and its treatment can be thought of as Corporate Social Responsibility.

Recommendation

The approach of private practitioners should be different when they treat a case of TB. TB treatment requires a higher cost. Private Practitioners should know the approximate cost of TB treatment; they should inform their patients regarding the approximate cost and duration of treatment and should also be concerned about the patient’s affordability to pay for the total cost of treatment. They should also give the patient a choice of getting treatment from the public sector if the patient is not able to afford the cost of taking treatment from a private hospital.

Migrants are living in their groups in the specific areas in the city. Mainly they are working in two specific industries: textile and diamond. For this group, strategies like screening at workplace and Basti approach (residential area) with a mobile van at their suitable time should be implemented. Screening of women should be linked through grass root level healthcare workers, and accompanying referral to the TB unit might become helpful in creating support to them.

The current study also suggests putting emphasis on IEC activities related to the benefits and services offered by the National Tuberculosis Elimination Programme so that the majority of the patients can be enrolled directly into the system and thus OOP expenditure can be minimized.

Financial support and sponsorship

The proposal was sanctioned by the RNTCP State Committee (Gujarat) for funding of the Operational Research.

Conflicts of interest

There are no conflicts of interest.