Portal de Periódicos da CAPES

Tuberculosis patient expenditure on drugs and tests in subsidised, public services in China: a descriptive study

2009; Wiley; Linguagem: Inglês

10.1111/j.1365-3156.2009.02427.x

1365-3156

Yanling Liu, Helen Smith, Yang Wang, Shenglan Tang, Qingliang Wang, Paul Garner,

Healthcare Systems and Reforms

Objective To measure patient expenditure on additional drugs and tests in public services where tuberculosis (TB) drugs are supplied for free. Methods Questionnaire survey of patients currently on treatment in eight TB dispensaries in two provinces; in depth interviews with providers at the facilities. Results Liver protection drugs (141/163) were prescribed for 86% of patients, 93% had one or more tests in the last month, and 23% were on treatment past the World Health Organization-recommended treatment duration. All but two patients were charged for something in cash at each visit: on average 287 Chinese Yuan (40 US dollars) in the previous month. For patients below the poverty line, drug and test expenditure was 1.85 times their average household monthly income. Average charges varied little between income categories. In terms of anti-TB drugs prescribed, 17.8% of regimens were inadequate by international standards. Providers reported they prescribe liver protection drugs to avoid medical negligence, and they believed they were effective; and the government subsidy for providing TB treatment was simply not enough. Conclusions Despite an ostensibly fully subsidised TB programme in China, patients are charged substantive amounts irrespective of income. Research is needed to confirm these practices are widespread but this needs to be coupled with financing strategies to tackle it. Objectif: Mesurer les dépenses des patients pour les médicaments et les tests supplémentaires dans les services publics où les médicaments antituberculeux sont fournis gratuitement. Méthode: Enquête par questionnaire auprès des patients en cours de traitement dans 8 dispensaires de tuberculose (TB) dans 2 provinces, entretiens approfondis avec les prestataires dans les services. Résultats: Des médicaments protecteurs du foie ont été prescrits à 86% des patients (141/163), 93% ont eu un ou plusieurs tests au cours du dernier mois et 23%étaient encore sous traitement au delà de la durée de traitement recommandée par l’OMS. Tous sauf 2 patients ont dû payer pour quelque chose en espèces à chaque visite: en moyenne 287 Yuan Chinois (40 dollars US) au cours du mois précédent. Pour les patients en dessous du seuil de pauvreté, les dépenses pour les médicaments et les tests étaient de 1,85 fois le revenu mensuel moyen du ménage. La moyenne des frais variait peu entre les catégories de revenus. En termes de médicaments antituberculeux prescrits, 17,8% des traitements étaient inadéquats par rapport aux normes internationales. Les prestataires de santé ont indiqué qu’ils prescrivaient des médicaments protecteurs du foie afin d’éviter la négligence médicale et ils pensaient qu’ils étaient efficaces. La subvention du gouvernement pour procurer le traitement de la TB n’était tout simplement pas suffisante. Conclusions: Malgré un programme TB entièrement et ostensiblement subventionné en Chine, les patients doivent s’acquitter de sommes substantielles indépendamment de leurs revenus. Des recherches sont nécessaires pour confirmer que ces pratiques sont répandues, mais cela doit être couplé avec des stratégies de financement pour y remédier. Objetivo: Medir el gasto de los pacientes en medicamentos adicionales y pruebas dentro de un servicio público en el que los medicamentos para la TB son gratuitos. Métodos: Encuesta mediante cuestionario a pacientes en tratamiento para la TB en 8 dispensarios y 2 provincias; entrevistas en profundidad con proveedores. Resultados: Un 86% de los pacientes recibió una prescripción de medicamentos hepatoprotectores (141/163), 93% había tenido una o más pruebas en el último mes, y un 23% estaba recibiendo tratamiento más tiempo del recomendado por la OMS. A todos, excepto dos pacientes, se les cobró algo en efectivo durante cada visita: en promedio 287 Yuan chinos (40 US dólares) en el mes anterior a la encuesta. El gasto en pruebas y medicamentos para los pacientes por debajo de la línea de pobreza, fue 1.85 veces el ingreso mensual promedio de sus hogares. Los cargos promedio variaron poco entre las categorías de ingresos. En términos de los medicamentos antituberculosos prescritos, 17.8% de los regimenes eran inadecuados según estándares internacionales. Los proveedores reportaron que prescribieron medicamentos hepatoprotectores para evitar negligencia médica, y que creían que eran efectivos; y que el subsidio gubernamental para proveer el tratamiento anti-tuberculoso simplemente no era suficiente. Conclusiones: A pesar de existir en China un programa de TB totalmente subsidiado, se cobra a los pacientes cantidades sustanciales independientemente de sus ingresos. Se requiere investigar más a fondo para confirmar que estas prácticas son comunes, pero ello debería ir acompañado de estrategias financieras para solucionarlo. The government in China has been tackling tuberculosis (TB) for more than 50 years (Wang et al. 2007). Since 2000, renewed government commitment has been matched with additional financial support from the Global Fund, World Bank (WB) and bilateral aid donors. The national TB control guidelines include free anti-TB drugs for 6 months to new TB patients, and for 8 months to retreatment patients. Tests include one free X-ray test for a TB suspect at first visit, one free sputum smear test for suspects whose X-ray tests are abnormal, and one free sputum smear test for patients under free anti-TB treatment, and providers are given a subsidy to provide the drugs. However, there is evidence that patients still pay (Table 1) (Meng et al. 2004; Zhan et al. 2004; Munro et al. 2007; Zhang et al. 2007). Even in provinces where TB drugs are free, patients are charged for drugs (Liu et al. 2007). One study stratified expenses by the income level of the patient (if costs vary with income, the burden is spread; but if costs are fixed, lower income groups are particularly disadvantaged). The estimated expenditure was 35% of monthly household income in lower income groups (Liu et al. 2007) and 17% in the higher income group, but the denominator and cut-offs were not given. There has been little research on what the charges are made up of Xu et al. (2006) found that health facilities in the study areas charge TB patients to increase their income through excessive tests. Hu et al. (2008) discovered that many patients are prescribed ‘liver protection’ drugs, for which there is no research or rationale for use (Liu et al. 2008). This is currently policy relevant. Pilot programmes to reduce these additional charges are being considered, with patients being reimbursed through the New Medical Insurance System for liver protection drugs; and the WB and donor supported programmes supplying free anti-TB drugs are being reviewed. Our study aimed to examine how much patients were paying for drugs and tests and relate this to their income. The patient survey was conducted in Chongqing and Sichuan provinces in southwest China, with populations of >30 million and 87 million, respectively. According to a nationwide TB epidemiological survey in 2000 (Technical director group for nationwide TB epidemiological survey in 2000 2002), the prevalence of active TB was 549/100 000 in Chongqing and 544/100 000 in Sichuan, which is higher than the national average of 367/100 000, and also higher than the rate in the mid-west region of China (451/100 000). We selected four counties from each province to represent the entire provinces, in terms of socio-economic development, geographic and transportation conditions (Table 2). We first classified all counties and districts in both provinces into eight levels in terms of per capita Gross Domestic Product (GDP), then considered the balance of terrain (urban or rural, hilly or plain, Chongqing or Sichuan), and finally selected one county/district from each economic level. Each county has a TB dispensary, and we collected data in these eight selected TB dispensaries. This resulted in us selecting, in the category up to 15 000 Yuan GDP, four of 42 counties; and for the category 15 000 Yuan GDP, four of 22 counties. Thus, if anything, slightly richer counties were overrepresented in the sample. All pulmonary TB patients currently on treatment were enrolled in the study as they attended health facilities for treatment. This included new patients (defined as patients who had not been previously treated for TB) and retreatment patients (defined as those who were previously treated for TB whose treatment failed, who defaulted, or who relapsed). The initial intention was simply to estimate across all patients the percentage of costs on TB drugs in relation to monthly family income. Sample size calculations were based on an overall estimate of the total cost of treatment across all income categories. Using health staff for surveys in China is common but introduces substantive interviewer and respondent bias. We therefore carefully trained and supervised four post-graduate students to interview patients, who collected data over 1 month. We intercepted the patients under anti-TB treatment as they finished their consultation and left the TB dispensary and continued to recruit on a daily basis until the required number of patients was interviewed. We asked patients about drugs and tests prescribed in the last month and the cost of these, as well as basic data on income, age and sex. Research supervisors checked the questionnaires as soon as they were completed and interviewed the interviewers to make sure all the items in the questionnaire were correctly understood and filled in. The data were double entered by two research supervisors. We categorised various drugs and tests reported by the participants and presented this data by new and retreated patients and against standard regimens. We categorised anti-TB drugs according to World Health Organization (WHO) guidelines and definitions of first and second line drugs (WHO 2003, 2006). We categorised liver protection drugs using a modified version of WHO standard definitions (World Health Organization Division of Traditional Medicine 2000) to be used in the evaluation and research of herbal medicines, which we also used in a systematic review of liver protection drugs (Liu et al. 2008). Ancillary drugs were categorised according to their functions, comprising anti-inflammatory drugs, drugs for symptomatic treatment (to relieve cough, asthma or haemoptysis) and drugs to reinforce patients’ immunity. We classed regimens as ‘adequate’ if they met WHO criteria for first or second line treatment. We classed regimens as ‘inadequate’ if they did not meet internationally accepted criteria. This included prescribing single-drug therapy; double therapy in intensive phase; second line anti-TB drugs for non-drug-resistant patients; Chinese medicines only and different combinations of WHO-recommended drugs, second line drugs, unclassified anti-TB drugs and Chinese medicines. Using the reported income data, we categorised patients into three levels of monthly family income: below the poverty line, below national average but above the poverty line and above national average. Income levels for the general population are available from the national economic and social development survey conducted by the National Bureau of Statistics of China; we used the most recent data from ‘Statistical Communiqué of the People’s Republic of China on the 2007 National Economic and Social Development’ (National Bureau of Statistics of China 2008). Based on the Statistical Communiqué, poverty line and national average income were 1067 Yuan and 8963 Yuan per capita per year, respectively. We divided these by 12 months to obtain income per capita per month, and then calculated the poverty line and average income per family per month by multiplying by 3.48, the mean family size in our sample. The levels for poverty line and national average of monthly family income were estimated to be 310 and 2600 Yuan, respectively. We tabulated income category against costs. We interviewed 11 TB medical practitioners from all eight TB dispensaries in two provinces. After providing verbal informed consent, semi-structured interviews lasting 20–40 min were conducted in Mandarin by the principal investigator, an assistant took notes, and the sessions were recorded. Standard topic guides were used to ensure relevant areas were covered during interviews. We used the Framework approach (Ritchie & Lewis 2003) to analyse the qualitative data and identify themes and managed the data in MAXqda software (VERBI GmbH, Marburg, Germany) using Chinese characters as described elsewhere (Smith et al. 2008). Ethical approval for the study was provided by Chongqing Medical University. We enrolled 180 patients, of whom 17 declined to participate: 137 of the 163 patients seen were new, and 26 were retreatment patients. Sixty-eight percentage were men, the average age was 42 years; 81% had a family income below the national average. Overall, the mean monthly family income of the study population was 1612 Yuan (SD 244.8, n = 163). Health insurance coverage was relatively low at 61% (Table 3). Treatment of 23% of both new and retreatment patients was extended beyond the standard treatment periods of 6 and 8 months, respectively, and patients had to pay for drugs prescribed during these extensions (Table 4). Of new patients, two-thirds were prescribed the standard TB drug regimen (provided to practitioners for free), 17.5% were given other standard regimens, and 15.3% were prescribed inadequate drugs or regimens. Of retreatment patients, 34.6% were prescribed non-standard and 30.6% inadequate regimens. Overall, 17.8% of treatments were inadequate, as defined in our Methods (Table 5). Liver protection drugs were prescribed to 86% of patients, mostly handmade herbal products (Table 6); we identified 25 types of liver protection drugs, which we grouped into four categories. None of the TB patients showed abnormal liver function. Liver protection drugs are prescribed to prevent possible liver damage induced by anti-TB drugs. Because no obvious difference between new and retreatment patients was evident, the data are presented combined. Ancillary drugs were prescribed to 35% of all patients, including anti-inflammatory drugs, drugs for symptomatic treatment (to relieve cough, asthma or haemoptysis) and drugs to reinforce patients’ immunity; we identified 30 types of ancillary drugs. Two-thirds of all patients had undergone two or more tests in addition to those provided for free in the previous month; additional X-rays had been taken of 68% of patients in the previous month. Less than 10% of patients had not undergone any test in the previous month. Only, two of the 163 patients were not charged any money. Patients were charged for liver protection drugs (86%, n = 141), ancillary drugs (35%, n = 57), tests (92%, n = 151) and TB drugs (36%, n = 59). This includes patients on extended anti-TB treatment who had to pay for their drugs in the extended period. The mean cost to patients of various drugs and tests prescribed during the previous month was 286.9 Yuan; spending on liver protection drugs comprised more than 43% of this cost. Total average expenditure on drugs and tests was relatively consistent between the three income groups (Table 6). For patients below the poverty line, drug and test expenditure alone comprised 1.85 times their average household income. We identified four themes from the analysis of interviews conducted with TB medical practitioners. Medical practitioners acknowledge charges. Most practitioners acknowledged they charged patients and that the financial burden of TB treatment is marked. However, they reported that they prescribed liver protection drugs and tests according to a patient’s financial status; they avoid prescribing liver protection drugs, or prescribe cheaper drugs, to patients who cannot afford them. A few thought expensive liver drugs were more effective. Medical negligence a perceived threat. Most practitioners reported that they prescribe liver protection drugs to ‘safeguard’ themselves from medical disputes. Many reported that the relationship between them and their patients was ‘tense’; of particular concern was fear of being blamed by patients who experience complications of treatment or liver damage. To ensure they are not held responsible for liver damage when a patient refuses liver protection drugs, several doctors told us they ask patients to sign their medical records to state they have refused the drugs. Practitioners believe in liver protection drugs. Most practitioners said that they had not studied the effects of liver protection drugs, but believed they worked, some saying that the Chinese population needed them to help tolerate Western drugs, and because anti-TB drugs cause liver damage and incidence of liver damage is high. Other additional drug, they believe, boost patients’ immune systems or treat patients’ complications and symptoms and can make patients feel better more quickly. One practitioner thought liver protection drugs were a psychotherapeutic placebo for patients but still thought it best to prescribe them. A few acknowledged the side effects of liver protection drugs. Government subsidies for managing tuberculosis patients are too low. Practitioners stated that subsidies for managing TB patients are too low in view of the workload. A few said they received subsidies for transfer, tracking and managing TB patients, but mostly these were village and township practitioners: many practitioners in TB dispensaries said they received no subsidies for managing TB patients, and they did the TB work mainly because of their conscience. Some said that other incentives are often earned based on performance. In China surveys are often performed by health staff but this introduces substantive interviewer and respondent bias. We wanted accurate data on financial burden against TB treatment costs, which providers would probably hide. For this reason, data collection was performed by carefully supervised students rather than government staff. This limited the number of participants, counties and study sites. This study was facility based, and only county TB dispensaries were studied, but they are the main facility for patients with uncomplicated TB being treated as outpatients, so these patients probably did not have complicated disease, and there is no reason to believe they belong to a particularly poor group. As the performance at TB dispensaries in these provinces tends to be rather similar, we selected eight TB dispensaries as representative and illustrative of the situation across the provinces. We collected data from virtually all patients attending, and the findings were dramatic in relation to excessive prescribing. Blood and X-ray examination were recommended almost monthly to more than 70% of patients, and the majority were prescribed liver protection drugs for which there is no rationale in Western medical terms (Liu et al. 2008). This results in high charges to people below the poverty line. Overall, the charges do not seem to vary much with the income status of the patient. Previous studies in other provinces and counties of China have shown the financial burden to many TB patients (Meng et al. 2004; Zhan et al. 2004; Xu et al. 2006; Liu et al. 2007). However, none of these studies examined charging against income level, or disaggregated the various components of additional charging. Our study shows that the various costs are fairly consistent across income groups and that the burden is very high for low income groups in these two provinces (Table 7). Why are these patients being prescribed unnecessary drugs? Income generation is probably the main factor, yet there remains a belief that liver protection drugs prevent adverse effects; indeed, many practitioners perceived they would be regarded as negligent if they did not prescribe these drugs. Currently, the doctor–patient relationship is somewhat uneasy in China (Song et al. 2003; Chen et al. 2005). Practitioners claim they prescribe to avoid accusations of negligence: what is unclear is whether they truly believe this, or whether it is just a convenient custom that earns income. What is critical to the argument is that the prescribers were clearly dissatisfied with government subsidies for providing TB care (Xu et al. 2006; Hu et al. 2008). Potentially, charging could contribute to low completion rates of TB treatment in this region, as found in Chongqing (Hu et al. 2008). Thus charging, with probable effects on adherence, coupled with the finding that almost one-fifth of patients in this study had regimens that were inadequate or irrational, including single-drug therapy, will increase the chance of multidrug-resistant TB developing and spreading. Despite the small sample size, data like this can be of value. In health services in China, many widespread practices are not acknowledged or discussed. This can be true for differences between actual adherence rates and government statistics and issues around charging in the government sector. In this policy environment, small research studies that demonstrate, for example, high charges, may compel policy makers and others to acknowledge the issue and open a discussion. Policy makers then need to evaluate the phenomenon more widely, or do something about it. Because the economic reform, health service providers in China have increasingly relied on user fees, resulting in overprescriptions of drugs and tests. This is at odds with the global TB programme, the WB/DFID funded project, and donor pro-poor policies, which try to ensure that TB treatment is free. Further work to confirm these findings can be generalised across the country needs to be linked with financial strategies to mitigate them. This research is an output of the Effective Health Care Research Programme Consortium, funded by the UK Department of International Development for the benefit of developing countries. The views expressed are not necessarily those of DFID.