Perceived Corruption and Quality of Health Services in Mbeya Urban District, Mbeya Region Tanzania

This paper examines the effect of corruption on quality of public health services. Corruption, in Tanzania, is a national concern that upsets public health services. There is a public outcry that corruption is increasing in the public health sector while the government’s effort to combat the phenomenon is unimpressive. This poses worries on the quality of public health services. The current study adopted cross-sectional research design, and a total of 180 respondents were involved in the survey. The Mann Whitney U Test was used to compare differences between perceived quality of health services and respondents’ characteristics. Overall, 87.2% of the respondents perceived low quality of health services, and corruption affected quality of health services to a greater extent. Based on age, employment and wealth status, there was significant difference on reporting perceived quality of health services at 5% and 1% level of significance. In addition, respondents’ sex and employment status showed significant difference in reporting the extent to which corruption affected quality of health services at 5% level of significance. The paper draws out two conclusions: first, the quality of health services was low. Second, corruption exacerbated poor quality of health services. To that effect, interventions to combat corruption in the public health sector are needed so as to improve quality of health services.


Introduction
Corruption, defined as the abuse of entrusted power for private gain in the public sector (Transparency International, 2004), is not a new term in the literature, so is the effort to combat it. The phenomenon is rampant in Africa relative to other regions in the world. Some African countries including Benin, Ghana, Zambia, Kenya, Uganda, and Mali are corrupt than Tanzania (REPOA, 2006). This implies that corruption is one of the widespread concerns in Africa, a situation that needs attention to bring about social development. Corruption is a governance issue, that not only affects public health and health systems in general, but also efforts to bring about development at all levels. Therefore, borrowing ideas from Streefland (2008) and Bradshaw (n.d), this paper argues that investing on strategies to combat corruption on the public health sector promotes social development.
Literature shows that access to basic health care services in developing countries was expected to improve since the 1970s after the Alma-Ata Declaration on Primary Health Care (PHC) and Health for All. This declared health as one of the fundamentals of human rights in addition to quality of health care that is an important element in transforming health services for development (URT, 2003). Despite PHC and the reforms thereafter following the World Bank and International Monetary Fund interventions that emerged in the 1980s, Sub-Saharan Africa (SSA) is lagging behind of all regions in the world regarding quality of health services. For instance, Streefland (2008) and WHO (2014;2015) contend that deaths related to pregnancy complications are higher in the region compared to developing countries as a whole. Similar sources show that infant mortality is high while life expectancy at birth is low. The trends in life expectancy, for example, indicate that by 1990, life expectancy at birth was 50 years for Africa region and increased up to 58 years by 2012-2013. In South-East Asia, it was 59 in 1990, increased to 67 by 2012 and to 68 years in 2013. Comparatively, life expectancy in Europe was 72 years in 1990, increased to 76 years in 2012. Globally, life expectancy at birth was 64 years in 1990, increased to 70 in 2012 and 71 years in 2013. Women continue to live longer than men.

Concept of Quality of Health Services
The concept of quality of health services is vague because, on one hand, as adopted in this paper, it depends on users' expectations, needs and preferences (Itam & Adindu, 2012). On the other hand, the quality of health services means provision of staff and facilities. Others relate the concept to equity and compassion, or optimum clinical outcomes or the degree to which health services increase the likelihood of the desired health outcomes (Akter, 2010;URT, 2011). Literature recommends that the resources invested should correspond to the specified standards and desired health service results. This means that the quality of health services should meet the needs of those who need the service most at a lowest cost (Itam & Adindu, 2012). Concerns in the health sector are many, but the most critical one is quality of health services that is likely to be affected by corruption. The quality of health services varies with economic, technological, political, educational and socio-economic and cultural factors. Thus, excluding quality aspect in offering health services is unprofessional potentially dangerous.

Corruption and Health Services in Tanzania
The provision of health care in Tanzania has pre -and post -independence history. The pre-independence health services were predominantly urban-based with the exception of mission health facilities that catered for rural-remote areas. The post-independence period International Journal of Social Science Research ISSN 2327-5510 2017 experienced redistribution policies, which embarked on a large-scale programme for ensuring access to health facilities as close as possible to the entire populace (URT, 2007). However, there is a growing concern about perceived quality of health services due to lack of reliability and efficiency of the health service delivery platform, knowledge and competence of the health workers (Akter, 2010).
In recognizing the link between corruption, health and development, and also understanding why individuals are engaged in corruption (Ngware, 2005), Tanzania is committed to combat corruption. Some of the efforts reported by REPOA (2006) include establishment of the Permanent Commission of Enquiry in 1966 to check the abuse of powers by government officials and agencies; passing a Prevention of Corruption Act in 1971; and forming the Anti-corruption Squad in 1975. In 1996, the Warioba Commission was transformed to be the National Anti-corruption Strategy and Action Plan in addition to Good Governance Coordination Unit in the President's Office to oversee all activities aiming at combating corruption in the government system at all levels. The government also transformed the Permanent Commission of Enquiry into Commission for Human Rights and Good Governance in 2001. In addition, Tanzania in its Development Vision 2025 emphasizes a government system that is transparent, responsive and accountable. This implies that the country emphasizes a government system that is free from corruption of every kind. However, 66% of the Tanzanians report increased corruption levels in the country. Urban dwellers perceive increased corruption than rural dwellers. Men also perceive increased corruption than their women counterparts. Similarly, highly educated perceive increased corruption than the less educated (Aiko, 2015). Unexpectedly, 82% among those who are compelled to pay a bribe in order to access public services do not report to the authorities because of fear and ignorance on reporting procedures (Aiko, 2015).
In 2005, health workers were the third most corrupt after the police, judges and magistrates (REPOA, 2006). Corruption, in its different forms, is prominent in accessing treatment from public hospitals. Other services that record extensive corruption incidences include getting assistance from the courts, police, water, sanitation and electricity services (Aiko, 2015). Although citizens' perception of the government's efforts to combat corruption was reported fairly by 62% of the respondents in 2005, it was reported badly in 2014 by 58% of the respondents (REPOA, 2006;Aiko, 2015). It is apparent from the foregoing information that corruption is increasing in all sectors of the economy in the country. Nevertheless, health indicators seem to be not improving effectively, something that poses worries on quality of public health services offered at different levels. The preceding information also suggests that the effort to combat corruption in the health sector, among others, is ineffective mainly because the phenomenon is complicated and challenging particularly in the health sector where corruption touches lives of the people direct.
The United Republic of Tanzania in its Quality Improvement Framework in Health care (2011)(2012)(2013)(2014)(2015)(2016) acknowledges that corruption is unbridled in the health sector and is likely to be one of the major barriers in providing quality health services (URT, 2011). Other studies have pointed out that corruption is a major hindrance in the delivery of quality public health services particularly among the poor (URT, 2007;SIKIKA, 2010;URT, 2011;UN, 2016). ISSN 2327-5510 2017 Despite institutional and legal transformation to fight corruption, the phenomenon continues to challenge developmental objectives in various sectors particularly the health sector (LHRC, 2013). However, it is difficult to generalize the effects of corruption on quality of services offered to the public, because corruption varies based on workers' culture, context and sectors of the economy (Melgar et al., 2010). Understanding the quality of health services and effects of corruption on quality of public health services in Mbeya Urban District is important because corruption remains high in the district (Ngware, 2005;Sulley, 2012). The next sections of this paper elaborate the methodology used, present and discuss the results, conclude and provide recommendations for improvements.

Methodology
The study was conducted in Mbeya Urban District in Mbeya Region. The district was selected because it has been implementing Tanzania Family Health Project focusing on improving quality of health services (Atherton, 2013). At the same time, the district is reported to have highest incidence of corruption in the region and the country at large (Ngware, 2005;Sulley, 2012) and therefore creating worries on quality of health services. The study adopted cross-sectional research design that allows data to be collected at a single point in time (Bailey, 1998). Data were collected through household survey by using a household questionnaire.
A multi-stage sampling technique was adopted. The technique was done in two main stages. The first stage involved selection of wards based on evidence of community involvement in Tanzania Family Health Project. Thus, three wards were selected randomly, namely: Isyesye, Itezi and Uyole. Secondly, in each ward, two streets were randomly selected making a total of six streets. A random sample of 180 individuals was selected, 30 from each street. A sub sample of 30 respondents is a bare minimum sample, in heterogeneous populations, for statistical data analysis regardless of the population size (Bailey, 1994). Data were analyzed by using Statistical Package for Social Sciences (SPSS). Frequency distribution was computed. The Mann Whitney U Test was used to compare differences between overall perceived quality of health services and respondents' characteristics. This is one of the non-parametric tests suitable for ordinal data used in this paper. The same non-parametric test was used to compare differences between overall perceived effect of corruption on quality of health services and respondents' characteristics.
A summated index scale was used to measure perceived quality of health services among the respondents. This had 17 statements. Every respondent was asked to respond whether he/she strongly disagreed, disagreed, was undecided, agreed or strongly agreed on each item of the scale. The scores ranged from 1 for strongly disagree to 5 for strongly agree. Overall, 17 to 50 represented low quality of health services, 51 represented moderate quality, and 52 to 85 represented high quality. The responses were grouped into three categories to reduce repetition of words, strongly agree and agree were regrouped into agree; strongly disagree and disagree were regrouped into disagree while neutral was treated as a separate response. Reliability analysis was done so as to assess internal consistence of the scale. In this paper, perceived quality of health services showed acceptable internal consistency with a International Journal of Social Science Research ISSN 2327-5510 2017 Cronbach's alpha value of 0.817 (Pallant, 2007;Field, 2009).

Wealth Ranking
Household wealth was assessed and the sample was categorized into poor and non-poor. The paper takes a wealthier household as one that owns assets including: radio, bicycle, television, houses with iron sheet roofs, and houses with cement floors, houses with cement/burnt bricks walls, cattle, phone, motorcycle and vehicle. The formula used to quantify wealth was as follows: Where: WET = wealth index yij = number of an individual asset (radio, bicycle, television, houses with iron sheet roofs, and houses with cement floors, houses with cement/burnt bricks walls, cattle, phone, motorcycle and vehicle) Ymax = maximum number of that asset in the sample X = number of items considered as indicators for wealth n = sample size Based on the wealth index median of 13, respondents were categorized as poor or non-poor. Those below the median were taken as poor while those above the median were considered as non-poor. This was important so as to ascertain which social group perceived low quality of health services as well as greater effect of corruption on quality of health services using Mann Whitney U Test. Table 1 presents respondents' characteristics. The results show that more than half of the respondents (50.6%) were males. Slightly greater than two-fifths (45.6%) were in the 21 to 30 years age group and 28.9% aged between 31 and 40 years. A closer look on Table 1 informs that majority of the respondents aged between 21 and 40. Since 50.0% of the respondents were married, it is unquestionable that this age group had highest fertility rate and respondents were likely to visit health facilities frequently for their under five children because of their special health needs implying that the sample was relevant for the study.

Respondent's Characteristics
Table 1 also shows that 45.6% of the respondents were farmers. With regards to respondents' education level, 36.1% held secondary education followed by those who held primary education. The number of those who had no formal education was also high. This brings about worries in terms of health and sanitation and social development more generally. In addition, 69.4% of the respondents were poor and more than two-fifths received health services at the regional hospital probably because the study was conducted in urban areas. At the national level, 81.7% of the population has primary education (URT, 2014). The lower proportion of primary education level holders in this paper is explained by the fact that the study was conducted in urban areas.  Favoring the rich relative to the poor implies that the poor were at risk in terms of accessing health services. Wealth ranking showed that 69.4% of the respondents were poor (Table 1). Therefore, marginalizing the poor suggests marginalization of the majority in the study area and the country at large because the majority is poor in Tanzania. In terms of overall satisfaction to the health services Table 2 shows that 72.8% of the respondents were not satisfied implying that the services were of poor quality. Literature acknowledges that the overall patients' satisfaction on the quality of health services is low in developing countries (Ahmad et al., 2011). However, this is contrary to a study done at Muhimbili National Hospital in Dar es Salaam Tanzania, whereby a high proportion of patients were satisfied ISSN 2327-5510 2017 with quality of health services (Mwakisu, 2005), possibly because Muhimbili is a national hospital where more resources and commitment of the government to improve quality of health services including combating corruption, are directed relative to the regional and district hospitals and health centers. Generally, patients' satisfaction is related to quality of health services (Leonard, 2008), such that increasing patients' satisfaction level reflects improved quality of health services.

International Journal of Social Science Research
Apparently, 71.1% of the respondents reported that doctors did not take patients' privacy seriously. This also suggests poor quality of health services in the study area. Literature shows that patients' satisfaction of quality of health services is explained by several factors including staffs who are caring and helpful without marginalization of some social groups. Other factors include responsive doctors and nurses to the patients' needs, skilled workers, workers' readiness to work regardless of circumstances, workers' competence and efficiency, workers' adequate communication with the patients, among others (Andaleeb, 2001). Figure 1 presents overall perceived quality of health services. Perceived quality of health services were categorized in three levels as low, moderate and high. The results show that 87.2% of the respondents perceived low quality of health services. That means the overall perception of quality of health services was low. The low level of quality of health services is largely explained by factors related to corruption itemized in Table 2 in this paper. They include long waiting time, marginalizing the poor and not taking patients' privacy seriously. Others include doctors demanding informal payments, offering health services on friendly perspectives and workers pretending as offering required health services. The low quality of health services has also been reported in rural and urban areas in developing countries in Africa and Asia (Kumari et al., 2009;Batool, 2005;Baba, 2004;Andaleeb, 2001). This implies that the poor quality of health services is a prominent problem not only in Tanzania, but also in other developing countries.

Figure 1. Respondents' responses on overall quality of health services
Based on the respondents' characteristics, the Mann Whitney U Test results showed that perceptions of quality of health services differed significantly by age, employment and wealth status. The difference is that the old age, the poor and employees in the public health sector perceived low quality of health services than others. This can be interpreted that the International Journal of Social Science Research ISSN 2327-5510 2017 mentioned groups were more likely to rely on health services provided by public hospitals. For instance, the labour laws in Tanzania require public employees, and their dependants, to acquire membership in the National Health Insurance Fund (NHIF) in which employees offer monthly contribution direct from their salaries. This requires them to access health services from public hospitals and to some degree from private hospitals. Some private hospitals however do not accept NHIF cards. In that situation, public sector employees are required to rely mainly on public hospitals that are ineffective in providing quality health services because of many factors related to finance and workers' behaviour. The poor also rely on public services because of low charges compared to the charges offered by the private sector to which the quality of health services is also problematic. In addition, the results in this paper have shown that the poor are marginalized in the public hospitals largely because they are unable to offer informal payments or bribe to the doctors and nurses, and therefore likely to report poor quality of health services relative to the rich.
Based on the national health policy of Tanzania of 2003, the old age should be given free health services in public hospitals (URT, 2003). However, its implementation is challenging because in most cases, according to the public outcry, the elderly are sometimes requested to buy drugs from private shops that are possibly owned by the doctors and nurses working in public hospitals. This is a common phenomenon that has persisted for many years in developing countries, and which is explained not only by the inability of the governments to finance the health sector (Streefland, 2008), but also corruptions in the public hospitals. The results in this paper are comparable to an observation made by Pastory (2013) who reports that the quality of health services is sometimes determined by patient's age and wealth status.  Table 4 presents perceived effect of corruption on quality of health services. The results showed that respondents agreed in 7 out of 10 statements that were used to measure perceived effect of corruption on quality of health services. Based on the results of this paper, the effects of corruption on quality of health services are multifaceted. Specifically, corruption lowers quality of health services. This is because it causes lack of motivation to the medical workers mainly doctors and nurses until they are given tips. It also leads to lack of drugs, equipment and other facilities. As such, the cumulative effect includes delay of the services and improper treatment that ultimately increase chances of poor health services and death.

Perceived Effect of Corruption on Quality of Health Services
Based on the experience of using treatment from public health facilities in Tanzania and on the public outcry, public health facilities normally lack necessary drugs and equipment like Electrocardiogram (ECG). This is a medical test that detects cardiac abnormalities by measuring the electrical activity generated by the heart as it contracts. Similarly, among other things, public health facilities lack Brain CT-Scan, a diagnostic tool used to create detailed pictures of the brain. To that effect, doctors and nurses direct patients to access such services from private health facilities that always have the equipment. This situation poses worries associated to corruption in public health facilities in the country such that corruption exacerbates poor quality of health services. The results are in line with a study conducted by Adindu (2010) on poor availability of drugs and services in public health facilities.  Table 5 presents perceived effect of corruption on quality of health services based on respondents' characteristics. The result of the Man Whitney U Test showed that there was no significant difference between respondents' characteristics and perceived effect of corruption on quality of health services. These results can be interpreted that respondents reported almost similar responses with regard to perceived effect of corruption on quality of health  ISSN 2327-5510 2017 services regardless of their socio-economic characteristics.  Table 6 presents results on the extent to which corruption affect quality of health services.

Extent to Which Corruption Affects Quality of Health Services
The results show that corruption affects quality of health services to a greater extent. The results in Table 6 show significant difference in reporting perceived quality of health services at 5% level of significance based on respondents' sex. Men respondents reported that corruption affected quality of health services to a greater extent compared to their women counterparts. This can be interpreted that men respondents were victims of corruption compared to women because of their position at a household level and community level as well, regarding gender relations. In African societies, men are the heads of households responsible for meeting all household needs including paying for health services in public or private health facilities. Because of that men are aware of corruption occurring during health services delivery relative to women. The fact that respondents accessed health services from public health facilities it positioned men to reporting corruption compared to women counterparts.
Employment status also showed significant difference. This can be explained that those employed in the public or formal sectors reported that corruption affected quality of health services to a greater extent. Clearly, those employed in the public sector are educated and therefore likely to be aware of any form of corruption and the extent to which it affects the quality of health services. Literature shows that men, educated and urban dwellers report increased corruption in Tanzania relative to their counterparts (Aiko, 2015). ISSN 2327-5510 2017  Note: numbers in brackets are percentages.

Conclusions and Policy Recommendations
Based on the results, the paper concludes that the quality of health services in the study area was low. Secondly, corruption affects quality of health services to a greater extent. For instance, waiting for a long time to get health services affected health services negatively and therefore patients were not satisfied with the quality of health services offered. Moreover, the elderly, the poor and employees in the public sector perceived low quality of health services, and therefore, perception of quality of health services varied with socio-economic characteristics. Apparently, corruption leads to lack of drugs, equipment and other materials. It also leads to poor quality of health services and patient's condition got worse due to delay of the services. In addition, regarding respondents' characteristics there was no significant difference between perceived effect of corruption on the quality of health services and respondents' characteristics.
Based on the conclusions, the paper recommends that the indicators falling under quality of health services for instance waiting time, privacy, respect, satisfaction and attentiveness should be addressed to improve the quality of health services. Secondly, the government should strive to ensure that doctors and nurses do not marginalize other social groups like the poor and the elderly during health services provision as a way of motivating corruption from the patients. This means that the government should ensure that corruption in the public health facilities is reduced to a minimum by posing labour laws that motivate medical workers to abide to the required values. This can be achieved by setting long-term goals, mobilizing resources for the public health sector and improving salaries and working environment among medical workers. Improved salaries and working environment is likely to discourage medical workers demanding tips to the patients. Other government interventions to improve quality of health services include: ensuring that the doctors and nurses are responsive to the patients' needs, medical workers are well skilled or are competent, medical workers are ready to work regardless of circumstances. Medical workers should also be efficient and should communicate with the patients adequately. Creating awareness on workers' values is also important.