Study Setting

The study was done in Tamil Nadu one of the states in south India. India has a robust public health system which runs through decentralized state budget allocation. With the advent of the National Rural Health Mission, a flagship health system strengthening program of the government of India in 2005, the public health system received a fillip in terms of decentralization, better platforms for community engagement with health care, better accountability mechanisms and greater fund allocation. [12] Alongside this strong public health system there is also a powerful private sector in health care. Private health providers, who deliver health care services for a fee, are the highest contributors to health services in India. There is a growing corporate health industry in the metropolitan cities which provide international quality health services not only to the people in the country but also serve as hubs for health tourism. In addition to the private and public health systems there is a big network of unorganized, unqualified medical practitioners providing all levels of health care. [13] The overall health expenditure is about 4% of the Gross Domestic Product (GDP) and the government budget allocation for health care is less than 1% of the GDP. [14] The remaining health expenditure is largely out of pocket. This leads to significant impoverishment and catastrophic health expenditure is one of the commonest reasons for indebtedness in the country. [15].

The public health system is plagued by system inefficiencies, irregularities, corruption and irrational health practices. The private providers largely remain unregulated and there is high prevalence of irrational care and commoditization of health care.

Tamil Nadu is one of high performing states in India with respect to health indicators. It has one of the well-functioning models of health care in the country but still several pockets especially poor rural areas and migrant urban populations remain largely underserved. [16].

Ethics Statement

The Institution where this study originated has an Institutional Ethics Group which does preliminary review of protocols to decide about the nature of review that they need to go through. The study protocol was reviewed by the Institutional Ethics Group which recommended an expedited review process for the protocol as the ethical risks were minimal. It was approved by the Institutional Review Board and Ethical Committee. Verbal consent was obtained from the participants as most of the respondents were unable to read and write. This was approved by the Institutional Review Board and Ethics Committee. Documentation of the consent was done on a record which was signed by the interviewer and a neutral third party.

Sampling

To address the main objective of the study the sample were selected from marginalized communities as resource deprivation is most prominent in these communities. Thirty five in-depth interviews were conducted, fifteen among migrant construction workers in and around Chennai, a metropolitan city in Tamil Nadu and sixteen among residents of a rural area in Dharmapuri district of Tamil Nadu and four interviews among primary care doctors catering to the health needs of these participants.

The migrant construction laborers hail from various parts of India, largely from the north Indian states of Orissa, Uttar Pradesh and Jharkhand. They are marginalized because of language barriers, migrant status and poor living conditions. Health access to these migrant construction laborers is very poor. Dharmapuri is one of the districts of Tamil Nadu with poor health indicators. In the surveyed villages the people are agricultural laborers and marginal farmers. In many households the men and women of economically productive age group migrate outside the village to urban areas for work. Since Dharmapuri is a border district in the border between Tamil Nadu and its neighboring state Karnataka, public services are compromised. Though the state of Tamil Nadu has some of the best health indicators in the country, Dharmapuri is among the poor performing districts of Tamil Nadu.

Interviews

In depth interviews were selected as the methodology for collecting data as the procedure is the most appropriate to gain insights into individual life experiences, trust, and meaning ascribed to trust. [17] The findings are representative of the meanings and experiences of those interviewed. It is highly meaning centric and cannot be generalized to the population. A trained interviewer conducted all the interviews. The construction laborers were approached in their place of residence and the rural participants were spoken to during their visit to primary health centers in their respective villages for health care. The interviews in the rural area were conducted in Tamil language and the interviews with the migrant workers in Hindi. The interviewer started talking to the respondents about health and their perceptions of health. Then the interviewer led the interview towards illness, treatment seeking and choice of health facility. This was followed by discussion about trust in health care. The various aspects of trust, what makes people trust the doctors and what makes people lose trust were explored during the interviews. The interviews last between 30 to 45 minutes each, with some interviews extending up to 2 hours. The interviews were not recorded to prevent the respondents from becoming self-conscious, which often happens among marginalized communities who have an inherent mistrust for research. [18] Notes were taken during the interviews by the interviewer.

Coding and Analysis

QSR Nvivo software package version 7 was used for coding and analysis of the interviews. The primary researcher read the notes several times and picked three information dense interviews. These interviews were coded by the researcher. The codes were verified and validated by a second researcher after discussions. Following this a coding manual was prepared with detailed descriptions of the codes. The remaining interviews were coded using this manual by the primary researcher. Due to the lack of availability of researchers trained in qualitative data analysis in the team, a third coder could not be engaged. In order to ensure an unbiased third set of coding and to have a fresh re-look at the data set for new perspectives, the primary researcher re-opened the data after a gap of 1 month and redid the coding of the interviews. The differences between the initial coding and recoding were identified and discussed with the second researcher till a consensus was arrived at. The codes were then grouped together into meaningful themes. The identification of themes was largely grounded in the data. But the influences of previous themes present in literature on the identification of these themes cannot be precluded. These pre-existing themes are adequately discussed along with the presentation of the results of the study. The main themes and their interrelationships were assessed based on the interviews. After identifying the themes, the conceptual frameworks were developed and discussed between the primary and second researcher.

Reflexivity

The primary researcher is a medical doctor by training and so in some of the early interviews could have brought the bias into the interpretation of the results. For example, in some of the early interviews when negative behaviors of doctors and health personnel were pointed out, the primary researcher felt defensive and this could have expressed in his body language and influenced the way the interview went ahead. A reflexivity journal was maintained by the primary researcher in which he reflected on the way certain statements of the respondents were interpreted and the possible alternative interpretations. These were reflected upon and appropriately addressed as memos during the analysis.

Dimensions of Trust in Health Care

The dimensions of trust in health care described here emerged as themes during the analysis of the interviews.

Perceived Competence of the Doctor/Health Facility

Communities have their unique perceptions about competence of the doctor or the health facility. There are many bases for these competence judgments. In resource poor settings, the lesser health awareness, and lesser access to information leads to highly subjective assessments of competence as compared to resource-rich settings where the higher levels of health literacy and access to information helps people make objective assessments. Community assessment of perceived competence tends to be shared. These competence judgments are often informed by prior personal experiences, opinion of community leaders, and shared opinions of friends and relatives. A common theme that emerged in most rural interviews was that the community perceived that private health facilities are better equipped and qualified to handle serious health conditions compared to public health facilities. An elderly man from a rural area also mentioned that primary health centers in the village are more suitable for women’s problems such as pregnancy, child birth and childhood illnesses. This opinion reflects the high emphasis placed by the public health system on reproductive and child health. While describing the community judgments about trust in the different health facilities an elderly man in a rural area mentioned,

“if there is some simple problem like knee pain, back pain etc. it is best to go to the Primary Health Center. If there is some other problem we can go to Harur Private Hospital. Harur Private Hospital is a big hospital and all big (sic) doctors are there. So all serious problems can be solved there”

Doing laboratory tests to make the correct diagnosis is perceived as a competent medical maneuver. In the rural primary health care settings, laboratory tests are not routinely done to arrive at diagnoses because of resource constraints. Therefore doing tests is perceived as a mark of competence. It was obvious in the interviews that the patients did not particularly know what the tests were and why they were being performed. But the acts of being subject to some tests made them trust the physician and the health facility. A young woman who had high fever had come to the primary health center for treatment. During her interview she mentioned,

“This morning the nurse did blood test for me. The result came within 2 hours and the doctor saw the result and told me that I have typhoid fever. She also advised me to take some more injections and some new tablets. I am feeling much better now and feel confident that I will become better. Without blood tests they wouldn’t have found out my problem correctly”

Though the knowledge differential between the doctor and patient prevented the patients from making an objective assessment of the quality of their care, they had their respective experiences which helped them form opinions on the competence of the doctor. Some of these opinions are informed by shared judgments of the community, but some are based on their own personal experiences. An elderly man with knee pain who was very happy with the treatment in a primary health center mentioned that the Primary Health Centre provides very good medicines.

“My entire family gets treatment for any of their problems here (in the primary health center). I refer all my friends and family to come to this PHC for treatment. That is because they give good treatment. I came here for my knee pain and the doctor saw me very well and treated me well. My pain went off within a week. The drugs work very well and we become alright soon.”

These perceptions of competence of the community are a reflection of the level of trust that they place on the doctor and the health facility. Perception of good level of competence also meant fulfillment of the treatment expectations that the patients had. Thus fulfillment of treatment expectations and competence are closely interlinked dimensions of trust. It was also evident that sometimes despite poor behavioral attributes of the doctors high levels of perceived competence undermined these shortcomings and led to trust.

In previous studies from the resource rich settings, it has been seen that clinical competence is an important component of trust in health care. Goold et al, in their review of various factors affecting the doctor patient relationship in the US have reported that competence is not only an important aspect for trust building; it is also a fiduciary responsibility of the clinician. [19] In the Indian and other resource poor settings, the legal component is not much but this study reveals a significant role of perceived competence in trust building. Competence as a dimension of trust in the western context rests on two important pillars, avoiding medical errors and providing the best possible outcomes of health care. [6], [20], [21] In the current study perceived competence emerged as an important domain but the language in which it was constructed was different. In resource poor settings where public health care facilities struggle to provide even the basic levels of care, expectations are also low. Therefore rather than mentioning medical errors and best possible care in their judgments of competence, the community referred to performing lab tests, making correct diagnosis and giving appropriate medicines in terms of ability to achieve a positive treatment outcome. Unlike the resource rich settings where individuals make judgments about competence of the physician or the health facility, in this study context a large part of competence judgment derived from shared community opinions.

Assurance of treatment

Assurance emerged as an important dimension of trust. Most respondents related trust to assurance of treatment. They worded trust as the surety or guarantee of some treatment from the doctor. This was narrated as assurance that some basic form of treatment will be provided, assurance of treatment irrespective of ability to pay, and assurance that at any time of the day some kind of treatment will be provided. One of the rural respondents of the interview said,

“the PHC is like our home. Sometimes we like to eat outside in hotels. But finally we have to come back to our home for home-food. Like that we can go to any private hospital we want and spend any money we want. But if we want something that we can always depend on then we have to come to the Primary Health Center. Whatever happens we can always go there and some kind of treatment will be given. They will not turn us away for want of money”

This strongly brings about several emotional layers of trust. In resource poor settings eating out is valued as a luxury that is reserved for special occasions. Home food is given high value as a daily source of sustenance, even if it is not as tasty as the food in the hotels or as nutritious. In the cultural context value is attached to home food in terms of guaranteed availability and surety. He equates treatment in the PHC to home food and treatment in private hospitals to hotels and restaurants. In this context, it seems to reflect treatment assurance in a situation of economic deprivation. In some interviews with migrant construction workers the non-availability of the doctor at the timings that they get relieved from their work responsibilities in the evenings came out as an important factor which makes them lose trust in the health system. A woman in a construction site in the urban area said,

“the doctors are always available in the clinic throughout 24 hours. One of the doctors lives very close to the clinic. So if there is any emergency during any time of the day the doctors can come immediately and help the patient.”

Another strong narrative that emerged was that the Primary Health Centers provided some basic form of treatment when the patient goes there irrespective of their ability to pay. The private health facilities were reported to have turned away patients because they could not pay. In a migrant interview a man said,

“last month, one of my colleagues had an injury. We took him to the clinic in Erumavetti Palayam. But the nurses there asked us if we have money. When we said that we did not have money, they just put bandage and sent us away. They refused to put sutures. So we can never trust the doctors and nurses there”

These uncertainties lead to lesser trust in the health system. Thus financial reasons had a strong underpinning in the dimension of assurance. Financial aspects influence trust in resource rich settings also. In a study in the US, it was shown that fee for service indemnity patients had greater trust in their physician than fee for service managed care patients. [22] But due to the distinct difference in the model of health care in the US and India, the way this aspect of trust is articulated is in terms of treatment assurance irrespective of ability to pay.

Willingness to Accept Drawbacks in the Doctor

In most of the interviews it was evident that when a trusting doctor-patient relationship was established based on certain domains of trust, the community was willing to overlook or accept the pitfalls in the relationship. The willingness to overlook the pitfalls was reflected in the level of trust that the patients had on the doctor. Though the willingness to tolerate shortcomings may be seen as a consequence of trust, it is also a reflective indicator of the level of trust in which as the willingness to accept shortcomings changes, trust can be perceived to change. In a particular interview with a young woman in a rural area she mentioned about the private doctor practicing in her village and referred to him as a rude person. But she added that whenever she takes her child to this doctor his treatment works very well and so she is willing to accept the rude behavior as she trusts that the doctor intends it in a good way.

“If the doctor is good it doesn’t matter even if he is rude to us. Even if he scolds us we know very well that he is doing it for our own good. Therefore it is most important for the doctor to be knowledgeable and good.”

This tradeoff can be seen as a reflection of the level of trust instilled by competence judgments, and a level of treatment assurance. In resource rich situations where there is a choice of doctors, it is reported that about 1/3 of the patients have dissatisfying clinical encounters. Of these a small proportion report complaints. Twenty percent of these are for communication problems and 10% for perceived disrespect or rudeness. [23] But this is very different in resource poor settings. Here behavioral issues such as rudeness are accepted as a tradeoff for competence or other trust dimensions. In resource poor settings, high out of pocket cost of health care is considered as a major deterrent in health seeking. In many of the migrant interviews the fact that came out strongly was that despite high cost, people seek health care with doctors or health facilities in which they place trust. Thus willingness to spend money (which is a scarce resource in the context) on treatment at a particular health facility/doctor emerged as an important dimension of trust. A migrant construction worker said,

“If anybody who works here becomes sick, we go to the nearest private doctor in Guduvanchery. We like and respect that private doctor. He takes a lot of money. But we prefer that. Even if he takes a lot of money he checks us properly and gives us good and powerful medicines. These medicines help us get better. Even if we spend money, we have to become better. If we don’t become better we cannot work. If we don’t work then what is the use of leaving our homes and migrating here”

When the vulnerability of the situation increased, this willingness to accept the drawbacks also increased. This was evident from a rural interview with an elderly man with diabetes. This man expressed high level of concern about his illness and seemed to be worried about its consequences as he was the sole breadwinner of this family and their economic condition was poor. When talking about his diabetes doctor in the nearby town he mentioned that despite having to spend a lot of money he feel comfortable because he trusts the doctor to make him better.

“The medications cost about Rs. 600 per month. I have to literally struggle to make ends meet in my house. Sometimes we even go without any good food to eat. But that is alright. Even if I have to work hard to make that extra money I don’t mind because I know very well that he is a good doctor and it is good for my health if I take the medicines.”

Loyalty

Trust was expressed as a sense of loyalty in some of the interviews. When the respondents spoke about some of the doctors or health facilities that they trusted the most, they referred to the fact that whatever the illness may be, they would come to the particular doctor or health facility. They also mentioned that they would not take any other treatment from a higher center without first consulting the particular doctor. This level of loyalty to the doctor is strongly influenced by positive experiences, comfort with the doctor/facility, perceived competence of the doctor/facility. It is also related to the willingness of the patient to accept drawbacks in the treatment given that all other domains of the trust are strong. A previous empirical study showed with the help of structural equation modeling that patient trust is an important predictor of interpersonal relationship and loyalty to the physician. [24] Thus loyalty is an important reflective dimension of trust.

Respect

Trust was also described as a deep sense of respect for the doctor. In one of the migrant interviews a woman said,

“the doctor is a very learned man. He has studied for many years to learn this noble art of curing people. I respect the fact that he is more educated and knows a lot of things about disease and treatment”

The patients perceived the differential in the level of knowledge between them and the doctor and respected the doctor for their knowledge. In addition the other aspects of trust also could increase the level of respect.

Determinants of Trust in Health Care

The interviews apart from giving an idea about the various dimensions of trust in health care, also gave insights on factors that determined the levels of trust. While perceived competence of the doctor, assurance of treatment, willingness to accept drawbacks in the doctor, loyalty and respect were various dimensions which reflected the trust in the doctor, there were some factors which were identified as determinants of trust.

Conclusions

This qualitative exploration gives directions for possible domains to be used for development of scales to measure trust in resource poor settings. Development and validation of scales to measure trust in health care in developing country settings would be an important direction to take. This can strengthen the move towards universal good quality health access to all in the developing countries by adequately informing the levels of trust in health care which will influence utilization of the health services.