Drug-resistant tuberculosis control in China: progress and challenges

China has the second highest caseload of MDR-TB in the world 1]. According to the National Survey of Drug-Resistant TB in 2007, one third of new
TB cases and one half of previously treated cases exhibited anti-TB drug resistance.
Some 5.7 % (range 4.6–7.1 %) of new cases and 25.6 % (range 21.7–30.0 %) of previously
treated cases developed MDR-TB 9]. The results of this survey confirmed an estimate of around 100,000 emerging MDR-TB
cases annually in China. Among MDR-TB patients, 7.2 % (range 4.9–10.2 %) were diagnosed
as XDR-TB, around 8,200 cases 9]. In addition, 11 % of new cases and 16 % of previously treated cases were resistant
to either isoniazid or rifampicin and were at high risk of developing MDR-TB 10]. Likewise, one third of patients with MDR-TB had drug resistance to either ofloxacin
or kanamycin 10] and were one step away from XDR-TB. These findings have sounded an alarm that the
prevalence of MDR-TB and XDR-TB in China could easily increase.

Although knowledge gaps as to the causes of MDR-TB remain, interrupted and/or inappropriate
TB treatments have been identified as the most important contributory factors in China
10]–12]. It has been argued that they can be at least partially attributed to health system
failures, in particular the reliance on a fee-for-services approach to financing public
health facilities. TB has been seen as a disease of the poor. It is not surprising
that the risk of developing MDR-TB is highest for poor and/or vulnerable members of
the population. Most cases are found in the less developed northeast and southwest
regions 13]. Some 80 % of MDR-TB patients are from rural areas, and most had low education and
were in the young to middle age-groups 14].

In the 1990s, the national TB control programme required the prompt referral of TB
suspects who had a cough for more than three weeks, hemoptysis or blood in a sputum
sample to TB dispensaries for diagnosis, and provided free first-line anti-TB drugs
for diagnosed patients. This was seen as essential in order to remove, or at least
substantially decrease, financial barriers to accessing standard TB care by the poor.
However, according to an evaluation of China TB control in 2004–2005, 70 % of suspects
were not referred to TB dispensaries for diagnosis 15] but treated as non-TB cases in public hospitals. Most patients spent more than half
of their annual income on treatment before being diagnosed with TB 16]. In addition, many studies in China have reported that TB patients are charged for
longer treatment periods than recommended by the TB control programme and that drugs
and tests are administered to an extent considerably beyond that specified in the
standard treatment regimen 17], 18]. As a result, many patients have paid excessive fees for TB treatment or have dropped
out and/or discontinued treatment because of difficulties in affording care 19]. In addition, a lack of proper training has often resulted in irrational prescriptions
and treatments 12].

Treatment of MDR-TB is complicated, expensive and often unsuccessful, resulting in
a low cure rate, high mortality rate and low follow-up rate 20]. The WHO guidelines for MDR-TB recommend 18–24 months of chemotherapy using a combination
of first- and second-line drugs (including daily injections in the first 6–8 months).
One systematic review that identified four studies on the cost of MDR-TB treatment
found that the cost per case was substantially higher in two locations where routine
care included substantial hospitalization (US$ 14,657 in Tomsk and US$ 10,880 in Estonia)
than where it only involved ambulatory care (US$ 3,613 in the Philippines and US$
2,423 in Peru) 21]. One study in China reported average daily medical costs for MDR-TB treatment were
more than three times the average cost of household daily non-food consumption in
Tianjin city and one and a half times in Henan province 22]. Some 92 % of MDR-TB patients in Tianjin and 70 % of patients in Henan experienced
catastrophic healthcare payments (defined in this study as daily medical costs being
over 40 % of daily non-food consumption) 22].

Progress on MDR-TB control in China

Drug-resistant TB prevention and control, especially with regards to MDR-TB, has been
an important component of the national TB control programme in China 19]. Since the 1990s, China has adopted the DOTS strategy for systematic management of
TB cases, implemented in a semi-vertical TB control system affiliated with the Center
for Disease Control and Prevention (CDC) at four levels: national, provincial, prefectural,
and county/district. National and provincial TB prevention and control centers were
usually responsible for programme administration, TB care supervision and case reporting.
Lower level TB dispensaries focused on the diagnosis of TB suspects, treatment and
management. TB patients with complications were referred to public hospitals. In 1998,
with the increasing prevalence of MDR-TB and other TB related diseases (e.g., TB/HIV,
TB/diabetes co-infections etc.), pilot exercises to integrate TB care into public
hospitals was undertaken in Shanghai, Jiangsu and Zhejiang provinces and a few sites
in the less developed western areas. TB clinics were set up within public hospitals
to provide TB/MDR-TB diagnosis and treatment and to cooperate with the local TB dispensary
on case management and reporting. This so called ‘designated hospital model’ for TB
control 23] is now being scaled up.

The central government provided annual earmarked funding to cover the costs of first-line
anti-TB drugs, two X-ray examinations and five sputum smear tests for TB, but there
was no designated fund for MDR-TB diagnosis and treatment at the national level 24]. Several international donors supported targeted interventions on MDR-TB prevention
and control in China, which were often project/programme based. For example, over
2010–2014 the Global Fund supported a programme on strengthening MDR-TB management
that was implemented in 89 prefectures of 30 provinces 25]. The main contents of this programme included: drug susceptibility testing (DST)
with smear-positive patients; DR-TB surveillance in project sites; the introduction
of rapid MDR-TB diagnosis technology; covering the cost of hospital admission for
MDR-TB treatment; providing MDR-TB patients with a transport subsidy; improving the
quality of second-line drugs produced in China; and ensuring a consistent supply of
second-line drugs. Matching funds from both provincial and prefecture levels were
required to co-financing the programme 25]. In the project sites of the Global Fund, 62 % of registered TB cases were tested
for drug susceptibility in 2010. The ratio of MDR-TB cases diagnosed to enrolments
on MDR-TB treatment was 57 % and over two thirds of MDR-TB patient sputum culture
examinations were negative by the end of 6-months of treatment, which was much higher
than in non-project sites 26]. However, the Global Fund programme is time-limited and sustainable long-term interventions
need to be established.

China’s ongoing health system reform towards universal healthcare coverage provides
important opportunities to improve access to appropriate care to prevent, diagnose
and treat TB/MDR-TB and protect patients from financial hardship. In 2009, a comprehensive
programme that aims to improve MDR-TB diagnosis and the quality and affordability
of treatment was developed and implemented in four cities, supported by the Government
of China and the Melinda and Bill Gates Foundation. The strategies included: the introduction
of rapid molecular diagnosis for isoniazid and rifampicin susceptibility and associated
laboratory staff training; standardized and detailed MDR-TB treatment protocols based
on the estimated degree of drug resistance; the use of health insurance and other
funds to cover 90 % of the medical costs of MDR-TB diagnosis and treatment and to
subsidize transport and nutritional supplementation; and strengthening MDR-TB patient
management using a TB control network, particularly in primary care facilities and
the community. The effects of the programme have been impressive, with a substantial
increase in the number of diagnosed patients and use of appropriate treatment. There
has also been a significant decrease in the average ratio of out-of-pocket payment
to annual household income 27].

In 2012 China CDC and the Foundation initiated Phase II of the programme. In addition
to confirmed effective diagnosis, treatment and patient management approaches, this
emphasizes a sustainable financing mechanism for TB/MDR-TB treatment. It was proposed
that health insurance schemes (NCMS, URBMI and UEBMI) should cover both inpatient
and outpatient TB care and that the reimbursement rate should be increased to 80 %
for TB treatment and 90 % for MDR-TB treatment. This was intended to reduce the financial
burden on TB/MDR-TB patients and improve patient adherence to treatment. A case-based
payment mechanism for TB/MDR-TB treatment was designed to ensure cost containment
and standardized clinical practice. In addition, the provision of a transport subsidy
for TB/MDR-TB patients was explored through cooperation with the Civil Affairs Bureau
which is in charge of the Medical Financial Assistance scheme. The strategies for
MDR-TB control in this programme are shown in Table 1. Health financing and payment reform is a critical component of the overall Chinese
health system reforms. Good practice and lessons drawn from this ongoing TB control
programme will be valuable for the ongoing development of health financing policy.

Table 1. Strategies for MDR-TB control in the China-Gates TB programme Phase II

Challenges on MDR-TB control in China and the way forward

Efforts to combat TB/MDR-TB in China have made substantial progress. However, critical
weaknesses could jeopardize effective implementation of the current strategy. In this
paper, we discuss challenges to TB/MDR-TB control in China from a health financing
perspective.

Although government funding for TB/MDR-TB control has increased year by year, there
remains a substantial shortfall in terms of providing the financial support needed
to ensure effective TB/MDR-TB diagnosis, treatment and management. Based on data from
the national population census in 2010, and the national TB and MDR-TB surveys in
2010 and 2007 respectively, an estimated CNY 13 million per annum is required for
TB control in each prefecture and CNY 4.6 million for MDR-TB 28]. A survey carried out in three prefectures in 2013 found that the annual funding
allocated to TB control was only one third of that required in two prefectures in
the eastern and central regions and much less in the western prefecture 24]. Funding for MDR-TB was far below the suggested level in all three prefectures, and
largely relied on the Global Fund, even in the developed eastern region 24]. When the Global Fund programme for MDR-TB control ends, the funding shortfall for
MDR-TB will increase substantially if domestic sources are not mobilized.

It is of great concern that the less developed regions often have a higher burden
of TB/MDR-TB but proportionally much lower expenditure on prevention and control.
This reflects the limited funds allocated to health services in general, which can
in turn be linked to their overall poor fiscal status. With the decline in international
donor funding, it has been proposed that low- and middle-income countries, especially
the emerging economies (e.g., Brazil, Russia, India, China and South Africa) should
increase the amount they spend on combating the MDR-TB epidemic as a means of ensuring
sustainable development. China’s economic success over the reform period has generated
the resources which would allow a substantial increase in the funding of initiatives
to address the MDR-TB public health crisis. These initiatives will need to take into
account regional disparities in both the challenges faced by those confronting TB/MDR-TB
and the overall financial resources available at the local level.

In China, MDR-TB diagnosis and treatment practices vary by prefecture. In general,
TB designated hospitals and the CDC at city and/or higher levels are able to conduct
DST and MDR-TB treatment. In some settings, county or district TB designated hospitals
and/or the CDC are responsible for delivering the samples from all smear-positive
patients and treatment failure patients to city or higher level facilities for sputum
culture examinations and DST. In some settings, smear-positive and treatment failure
patients are recommended to visit city or higher level facilities in person for DST.

According to the CDC TB register, the average delay in 2013 between identification
as an MDR-TB suspect and an MDR-TB diagnosis ranged from 59 to 83 days in the three
prefectures in the eastern, central and western regions identified above 24]. Interviews with the head of the TB department in the local CDC, TB designated hospital
managers and TB care providers in each of the three prefectures identified some common
constraints resulting in long delays in MDR-TB diagnosis. These included lack of funds
and/or incentives at county or district TB facilities to ensure the correct and timely
delivery of patient samples, and the perception by patients, especially poor patients,
that they would have difficulty in affording care 24].

After diagnosis, it is recommended that all MDR-TB patients should be admitted to
a prefecture TB designated hospital for an initial 1–2 month period and then should
continue outpatient treatment for 16–18 months. Treatment is individualized, being
determined by the specific drugs to which a patient exhibits resistance, and the cost
is some ten times that of standard TB treatment 28]. Although most rural patients have NCMS coverage, the reimbursement rate for hospital
admission is usually low (based on the NCMS principle of lower reimbursement rates
at higher level health facilities) and outpatient costs are often not covered. A survey
in three prefectures located in the eastern, central and western regions in 2013 reported
that the average out-of-pocket payment for MDR-TB treatment over a 24 month period
was CNY 20,544 24]. Non-medical costs (including fees for transport and accommodation during treatment)
were around one third of the medical cost. In this survey, almost all patients reported
that the financial burden on their households was heavy or very heavy and half reported
borrowing money from their relatives or a bank to pay for treatment. These three prefectures
were all sites covered by the Global Fund programme, which provided financial support
for treatment and transport. It may be inferred that MDR-TB treatment would be less
affordable in non-project sites and might give rise to more serious economic and social
consequences.

The China-Gates project Phase II introduced new financing and payment methods to reduce
the financial barriers to accessing MDR-TB treatment (including cooperation with the
Civil Affairs Bureau to subsidise patient travel costs). However, it was recognized
that without appropriate incentives to both TB designated hospitals and their TB care
providers, there was a risk of undermining this intervention. At the prefecture level,
the revenues of TB designated hospitals were still largely from service charges 24]. In qualitative interviews with prefecture health administrators, CDC directors and
TB designated hospital managers in the three prefectures, most expressed the opinion
that the implementation of case-based payments for TB/MDR-TB treatment in prefecture
TB designated hospitals would lead to a decline in hospital revenues 24]. This would reduce the incentive to treat TB/MDR-TB and might negatively impact the
quantity and/or quality of care provided. Another important consideration is that
the salaries of TB care providers in hospitals are directly related to the associated
service fees. Most TB care providers interviewed were not satisfied with their current
salaries, which were typically less than their colleagues in other departments. There
was also concern that they might be exposed to a high occupational risk in treating
infectious diseases. This had made it difficult to recruit TB doctors 24].

Financial and material incentives have been directed at individual healthcare providers
and organizations (both public and private) in many other low- and middle-income countries,
aiming to improve the quality of TB diagnosis and treatment, typically alongside system
changes intended to promote improved outcomes. For example, in Romania and Honduras
incentives (e.g., gift tickets or other materials) were given to public healthcare
providers conditional on objective performance indicators, for example the number
of new cases confirmed by microscopy, the rate of DOT in sputum-positive patients,
and patient attendance for TB treatment 29]. In India, the Philippines and Myanmar, private healthcare providers were supplied
free anti-TB drugs on condition that patients did not pay for these 29], 30]. Dispensing free drugs is seen as an incentive for private providers because they
can charge consultation fees and develop their reputation for curing TB patients,
which might raise client demand for other services. There have also been trials involving
the provision of conditional financial incentives at organization level (e.g., non-governmental
organizations, anti-TB teams, local governments) linked to TB control performance
indicators 29]. It is difficult to fully attribute performance changes to these incentives given
that they were often just one component of multifaceted interventions. However the
evidence does seem to indicate increased effectiveness in case detection and treatment
completion where incentives were offered to providers and/or patients 29]. Findings from studies in a number of countries emphasise the need for great care
in both the design of such incentives and their implementation, which requires a detailed
understanding of the environment within which providers operate and their needs, as
well as scientific evaluation of effectiveness. These international experiences should
be considered in the on-going reform of TB diagnosis and treatment financing in China.
Developing a hospital compensation strategy that includes appropriate incentives for
TB care providers will be an essential component of an effective intervention.