Background

Many Asian countries, including China, are going through profound socio-historical transitions due to continuing urbanization, marketization, and demographic changes. As these changes have occurred, it has become critical, and simultaneously difficult, to assess the needs of the elderly populations for a number of reasons. First, the transition of the service environment from a traditional family and neighborhood-based care support system to urban community, institution, and market-based eldercare service systems has typically increased the number of those whose needs are not being met as a result of the disintegration of the traditional care network and the lack of awareness of the service needs, and information of service availability during this transitional time. Second, the transition of lifestyle from a traditional agriculture-based one to an industry/technology based one represents a shift from a life focused on meeting basic physical and social needs to one with a broader spectrum of needs and subsequent re-prioritization. Third, the transition of political governance from a state-centered monopoly in countries such as China and Vietnam, which traditionally promotes collective consciousness and focuses on generic norm of needs, to an increasingly marketized society, which allows tremendously more local and individual diversity has resulted in a wider range of needs, many of which are being unfulfilled in this transitional period.

As indicated in a number of recent community-based survey and focus group discussions in China (e.g., Fu 2010; Lowry 2009), those elders who are in the middle of these multidimensional societal transitions are burdened by and uncertain about shifting social norms and expectations. More specifically, they have difficulty comprehending changes in social acceptance and changing priorities of certain needs, legitimized resources and methods for needs fulfillment, acceptable habits of needs expression, articulation, and requests for social support. When this new needs-related psychological complexity is superimposed on the substrate of common age-related changes in the physical, social, and mental landscape of an elderly person, accurate and authentic articulation of personal needs often becomes more challenging.

To address the needs of elders, a needs assessment, broadly defined as systematic process for establishing priorities and making decisions regarding program planning, development, and operations, must be conducted. In particular, a needs assessment must be conducted to determine if gaps exist between “what is” and “what should be” regarding elder support and services. To conduct a needs assessment, two commonly used methods are standardized questionnaire survey and qualitative interview. The standardized questionnaire is comprised of a series of questions and is used to gather information from respondents. The standard questionnaire’s advantage is its ability to be widely and easily distributed to the target population or region. In addition to the common challenge of tailoring the wording of the survey questions to elicit accurate responses from culturally-diverse respondents, low response rates of the questionnaire survey method results in the inability to understand why participants do not respond, as well as the inability to interpret atypical responses obtained. In light of these barriers, some researchers favor the qualitative interview method to perform the needs assessment (Dingwall et al 1998). The qualitative interview follows a semi-structured schedule to cover major topics in pre-determined domains, while allowing the flexibility of customization of the inquiry process necessary for capturing the most accurate information. The interview method, if not used correctly, can yield dubious results on elder and caregiver needs, just as poorly designed standardized questionnaires can.

Researchers have long noticed limitations of the qualitative research methods such as interview and observation. For years, the tension or gap between the emic perspective (as in interviewee) and the etic perspective (as in researcher) have been deemed challenging (Headland et al. 1990; Patton 2010) and can potentially be fruitful especially when we “demonstrate how participants’ perspectives may diverge dramatically from those held by outsiders” (Yin 2010). In the standard elderly care needs assessment such as the Camberwell Assessment of Need for the Elderly (CANE) (Reynolds et al. 2000), for instance, both self-reporting from the interviewee, and observation by the interviewer, were used to supplement each other. The challenges, however, remain not only in design of a needs assessment instrument but also execution, data analysis, and comprehension of the actual need scenarios, due to unbalanced emic vs. etic approaches in design and execution of interview.

This study summarizes the experience of pilot testing of the interview method used for assessing the needs among elderly people in urban China, with a particular focus on the limitations of the interview method and how to improve the process of inquiry.

Methods

Instrument Development

The interview guide was initially developed based on a review of several existing needs assessment scales, including Camberwell Assessment of Need in the Elderly (Reynolds et al. 2000; Orrell and Hancock 2004) that were originally developed from UK and the Strengths and Needs Assessment of Older Adults (Hayden et al. 2004) developed from the US as well as ones from China (Fu 2010), in order to capture a set of needs categories that are common among seniors, persons with a disability, and their family caregivers. The initial interview guide, which included three broader categories of physical, mental, and social needs, was reviewed by several stakeholders including two eldercare service managers, three senior service researchers, two elderly people, and three family caregivers of elderly persons. Their feedback, mainly focusing on inclusion of more specific categories of needs, together with editorial advice about wording of particular items, was incorporated in the revision of the final interview guide.

The final version of the interview guide for the senior needs assessment covers the following six areas of needs: (1) health care needs, (2) physical or survival needs, (3) security and financial needs, (4) psychological needs, (5) social needs, and (6) spiritual needs. Together, the assessment included over 36 questions. Some of the questions posed were, “Do you have health insurance?” Do you have severe depression, or other mental illness that limits daily activities? “Do you the constantly suffer from severe pain?”

Interview

The cases that are selected below for illustrative purpose in this study are part of the qualitative interview data collected from the elder care needs assessments conducted via home visits in the urban districts of Beijing and Shanghai in China. A total of 42 interviews were conducted through home visits conducted over a period of 8 months from the Fall of 2012 to the Spring of 2013.

Participants included: (1) family caregivers who provide care to family members with light-to-moderate Alzheimer’s, and (2) family caregivers who provide care to those with more advanced stages of Alzheimer’s. Each of these family caregivers were (i) age 18 years or older (either spouse or adult children caregiver), (ii) self reported as a caregiver for someone with significant memory problems or with a diagnosis of dementia, and (iii) self reported as a primary caregiver, specifically having been in the primary role of caregiver for an average of 4 h per day for the prior 6 months.

Participants also include service providers from community’s aging, health, and home service providers including those from aging services (e.g., senior center, nursing home), health services (e.g., primary care, geriatric/psychiatric/neurologic clinic), home services (e.g., home attendant agency). The main inclusion criterion was having 1 year or longer experience providing care, support, or assistance to caregiving families.

Finally, family members of elderly persons with no dementia or other critical medical conditions also participated in our study.

Results

Based on these experiences, we have observed a number of limitations of using the interview method to conduct needs assessment among elders. These limitations or problems can be grouped into three main categories.

  1. (1).

    Socio-cultural limitations in self-reporting: Informants, particularly those of Asian background, though they are fully aware of their needs, often tend to underreport them because of two reasons: (a) a sense of family loyalty, and (b) a sense of societal hierarchy. The following cases are illustrative of these problems.

Case #1 demonstrates how under-reporting can easily occur during the interview: An old man, who could hardly speak because of his Alzheimer’s disease condition, was taken care by his daughter-in-law and son, while his teenage granddaughter helped care for him after school. The daughter-in-law who did not have a job took care of the old man during the day, while the adult son took care of his father from evening until well into the night. The whole family warmly greeted us at our arrival during the scheduled evening time right after the meal. When we asked the daughter-in-law about examples of “caregiving situations that bother her the most, or [that] she feels are very difficult to deal with”, she smiled and gently replied, “I’m fine. I don’t have much to complain. We work together as a family. My husband and my daughter are all very supportive. I really cannot think of any situation of particular difficulty.” In addition, her husband stated that he didn’t mind staying up late to take care of his father.

On the notebook, the interviewer marked “no need” next to that item. However, in a follow-up visit, which happened to be a morning when the daughter-in-law was alone with the old man. This time, to our surprise, she poured out how stressed she had been feeling about the lack of time that she now had for everyone else in her life, especially her daughter who often needed to help with her grandfather and could not get her homework done at night. More seriously, she said that she really worried that her husband might commit suicide because he was under a lot of pressure from his work, but needed to take care of his father in the evening, when he was also expected to get his work done. Above all, she revealed that her father-in-law was a very difficult person, even before he became sick and became worse after he was diagnosed with dementia. This time, the interviewer jotted down a full page of problems and needs of this family. This case showed that interviews on elder caregiver needs should leave more space for open-ended questions about how things are going and to leave more time for the interview and to pay close attention to whom is present during the interview.

Case #2 shows how an interviewee may deliberately select what to report: A retired lady in a Shanghai neighborhood was taking care of her husband, who had a stroke a few years before and had been recovering ever since. At our first visit, an interviewer who was a graduate student from a local university talked to the old couple. When asked about the most difficult aspects of caregiving for an old man, the old lady mentioned little except for the old man’s moody personality—a trait that the older man had had even before the stroke. At the second visit, another interviewer who was from the U.S. but able to speak the local language well went together with the local colleague who had visited this family before. The old lady and man both expressed that the most difficult situations they encounter on a regular basis, especially during the winter season, was not having a place to take a bath. The older apartment buildings that were built in that region typically did not have a bathtub or a shower room besides the toilet and kitchen space. The old lady clearly indicated that she did not say this to the previous interviewer because she did not think that a junior student had the influence necessary to enact change. But this time, she (mistakenly) thought that the interviewer from the U.S. on account of his formal suit was from the upper level government, and therefore she revealed a true need with the hope that such an expression might lead to improvement in the future.

Because interviews in Asian countries are often administered in front of other family members, and the elderly in Asian often are reluctant to be truthful with other family members, a private setting should be considered. Additionally, because many elders grew up in an age where only those of advanced age had authority, informants need to be told of the changing nature of Asian society, where younger individuals have power as well. In Case #1, the informant seemed to have concerns about reporting the truth in front of other family members; it argues for a private setting, which could be hard to access physically and socially in crowded urban China, in order to obtain better outcomes. In Case #2, the informant deliberately chose to report the true need for a bathing device only when she determined that her audience might be powerful enough to influence desired changes.

This kind of under-reporting or misreporting due to social or cultural reasons has been extensively discussed by social linguists around a broader concept of performativity (or performative utterance) in that saying something often leads to effects, intended or not, beyond simply reporting on or describing reality (Parker and Sedgwick 1995). In this case, an informant may deliberately over-report if perceiving possibility of desirable outcome, or under-report if perceived fear and concerns of negative outcomes, or futility of leading to no desirable outcomes. Interviews are often not just about expressing inner truth but about performing for an audience of interviewer as perceived by the interviewee and for an audience of family and friends present at the time.

  1. (2).

    Cognitive limitations in self-reporting: People tend to under-report the needs of oneself or one’s relatives due to limitations in cognitive capacity or styles.

Case #3 shows the importance of having multiple informants when the interviewee is cognitively impaired: An old couple and their son lived together. During the interview, the son was the main person talking to the interviewer. He told us how he quit his job after receiving the news about his mother’s stroke, how he first accepted his role as a caregiver whole-heartedly to reciprocate all the kindness she used to give to him, and how this changed over time. Half a year into the care for his mother, he started to feel so bored with his life. He began to notice that he easily got angry, felt envious of other people of his age who can go to bars to drink beers, and he started to feel depressed. Nevertheless, he continued to take care of his mother. During the whole conversation, the interviewer tried to elicit some responses from the young man’s father and his sick mother who could barely speak due to stroke that was accidentally discovered by the interviewer at the end of the interview, but received very minimum answers.

On a follow-up visit, the son happened not to be at home, and a neighbor was in the house chatting with the old man, the son’s father. The neighbor, who was a long-time friend of the family, told the interviewer that the son was a headache for the old couple. The real story was that he was fired from his previous job. As a result, the old couple were very worried about him, their only child. The mother was so worried that she had a stroke. The father asked him to come home since he was unemployed. He came home and soon became agitated, full of anger and complaints, and spent most of his time outside the home at a friend’s house. His old father was the main caregiver, who needed to cook, wash, and do everything for both his wife and his son. Sometimes, the old man became very depressed and did not want to speak to anyone. “It’s very sad.” The neighbor said, “Sometimes I come in, I see the old couple, one in the bed, one sitting at the bed side, holding each other and crying. The house used to be so lively before the mother got sick. Now everything started going downhill.” In this case, the informant’s reporting seems influenced by not only the role as a caregiver but also the relationship with the patient with dementia. Thus, it is beneficial to have multiple informants in order to detect possible limitations of a particular informant.

Case #4 suggests that the informant may not report certain risks such as alcohol use as he or she may not know it is a health risk: Mrs. C took care of her dementia-afflicted husband for almost 10 years. Less than a year after his retirement, he started to drink heavily at home. He had enjoyed liquor all his life, but he drank more and more after retirement, probably just “because he had nothing else to do in life,” said his wife. He would have a glass in the morning right after he got up to lift his spirit, and have a few drinks at lunch, and a few drinks again before and after the dinner. After he showed signs of loss of memories, he drank even more. Mrs. C. again reported: “If you don’t give to him, he would get angry. So I usually let him drink, or do whatever he wants.” When asked whether he would get violent such as hitting her when he was angry, Mrs. C said never; he has been a very gentle and nice to her all his life. In general, he did not like to talk much. When asked whether he ever worried that his drinking too much would be a problem, she said that it did not occur to her and he was pretty healthy physically. They went to a doctor who basically said that there is no medicine to treat her husband’s problem, telling her that it’s like a slow cancer. So, just let him do whatever he enjoys to do. When the interviewer asked her whether her husband ever had a formal diagnosis, whether he currently takes any medicine, whether they had seen any other doctors for confirmation, and whether it ever occurred to her that drinking alcohol might be a risk factor for dementia, the answers are all “no.” The doctor she had seen, in fact both she and her husband had seen throughout the past two decades, is the only the doctor that she used, who is assigned to her based through the benefit package from her husband’s company. In this case, neither the patient nor the doctor asked about the risks for dementia.

Comments: In Case #3, the young man’s narcissistic thinking led him to a biased perspective on the household situation, while the couple did not report much due to either the impact of the stroke on the old lady or possibly the depressed mood in the old man. In Case #4, we see that a lower level of literacy might contribute to an incapability to articulate, and a lower health literacy might lead to both an inability to articulate the health care needs, and inaction in seeking health care. In this case, both Mrs. C and her husband had limited knowledge about dementia, and even their doctors seemed to have limited knowledge at least about possible risks for dementia. These are typical examples of tension between emic and etic perspectives. Some informants may not report a certain need and experience, because of limited literacy, limited health literacy, or cognitive disabilities due to dementia or stroke, even if they are partially aware of the situation; other informants may have a distorted perspective due to depression, personality disorders or other clinical reasons. Meanwhile, the researcher whose thinking is influenced by biomedical knowledge about dementia risks expects to hear reporting of the risks established in the health literature.

  1. (3).

    Conceptual gap: Discrepancies in perspective, knowledge, and beliefs can also become a barrier to articulating or even recognizing one’s own needs, especially needs for health care services.

Case #5 shows again that an interviewee’s perception of his or her own needs can be very different from that of an interviewer: A single, never married old woman lived by herself. When walking into her apartment, in addition to the messy surroundings of her living room, there was a pungent odor of feces in the air, which got stronger as we got closer to her. During the interview, she was not very talkative, responded primarily with a brief reply to all the queries. She reported her health issues including having arthritis and being over-weight. She still was able to move about and cook for herself, but she relied on her sister, who lives in the same town, to bring her groceries once or twice a week. She did not feel particularly lonely or depressed, because she already had gotten “used to it.” When asked whether she was able to do everything for herself, such as getting dressed, going to bathroom, and cleaning herself, she was very positive. Even when asked whether she had concern about hygiene, she clearly said no. Finally, the interviewer had to say directly that there seemed to be some odd scent in the room. Her response was a kind of surprise. She apologized and said that she would light an incense right after the guest left. The challenge to the needs assessment could be due to a loss of sensitivity of the olfactory system (of the informant), thus reducing the informant’s ability to notice the issue of sanitation. Partly, sanitation and order (or being unsanitary and messy) often is a quite subjective matter and the standards vary across sub-cultural or ethnic groups and individuals. How can the interviewer determine the needs in these aspects when the interviewee is not aware of such a need? If the interviewee denies such a need, and if an interviewer checks the “unsanitary” based on her observation, is the interviewer imposing her own personal or sub-group cultural standard?

Case #6 shows that an interviewer’s mind set, or conceptual framework, may interfere with the interview protocol: An interviewer, who was trained to conduct the semi-structured interview, was very interested in Maslow’s theory of needs hierarchy. When interviewing an older professor who had retired and stayed at home living with his daughter’s family, he systematically asked the old professor’s perceived needs starting from basic physical needs and went up through Maslow’s hierarchy of needs. The interviewer was eager to ask the questions regarding the old man’s perception of the needs for personal growth and self actualization, an issue held so dearly in the interviewer’s mind that he used to argue with his classmates that self actualization ought be the most important thing to have for an aged person, and perhaps the most desired form of life. When he heard the professor’s response of not having those needs, the interviewer couldn’t believe and accept his response, and asked more directly, “Have you heard of the words self actualization?” “I suppose I have, but I never really understood what it means. You know I am an engineering professor.” “Like you are happy with what you are doing, you are interested, and you find great meaning in it.” The old man said that he never liked to exaggerate the meaning of his work, “To me, working is to solve problem. Sometime you are happy with the results or process, and a lot of time you are not happy. But that was before retirement. Now I only want to enjoy taking care of my granddaughter. If she is happy, I am happy. If my daughter is happy, I am happy.” The interviewer was clearly in his own conceptual space, where the professor being interviewed shared neither Maslow’s view nor the interviewer’s value of Maslow’s hierarchy. Although it is legitimate for an interview to ask additional questions to confirm a response when needed, the interviewer’s preoccupation with a particular concept, which appeared similarly in all of his interviews, clearly challenged the efficiency of his communication with the interviewees. The question is when the interviewee’s perception of needs (or needs fulfillment) is not congruent with the interviewer’s conceptual framework, to what extent does the interviewer have the legitimacy to make a judgment of whether a certain thing, as an indication of the concept of the interviewer’s perspective, exists or not?

Comments: Both cases suggest the importance of maintaining a balance between the emic and etic perspectives throughout the interview process. In Case #5, the informant’s self-report (i.e., the emic perspective), which is based on her loss of sensitivity, should be supplemented by the independent observation that represents the etic perspective of the interviewer. Case #6 indicates that the interviewer’s strong etic position, though legitimate, could be further refined with sufficient training or self reflection on one’s own etic perspective. It is imaginable that the interviewer could have a rather inclusive view of self-actualization needs and accepted that the old professor’s enjoyment of his granddaughter’s joy is a joyful self transcendent experience of the prolonging of one’s self, based on an Asian view of Maslow’s concept.

General Discussion

The interview method is often considered advantageous for conducting a needs assessment because, unlike a survey which has fixed question, the interview can use a semi-structured guide to address key domains of the problem, combining with open-ended queries and customized flow to allow the interviewer to flexibly probe, explore, and reiterate for the purpose of deeper revelation, better clarification, and further interpretations. The results of this study, however, revealed some challenges to using the interview method to collect accurate data for assessing needs of elders and their families. Since it is often hard to determine whether the limitations are due to lack of an experienced interviewer or poor quality of an interview protocol, we believe that modification and improvement of both aspects can be beneficial.

The interviewer factors can be improved through adequate training and appropriate execution of the interview protocol. As Dingwall et al concluded, “qualitative research requires real skill, a combination of thought and practice and not a little patience”, suggesting the importance of improving the interviewing skills in order to increase the accuracy and validity of collection of interview data. A typical recommendation is to seek a balance between the emic approach and the etic approach in the interview. To strengthen the emic approach, the interviewer should spend sufficient time before the interview to allow the interviewer to get familiar with the interviewee and his or her living background, and to establish rapport with the informants and manage the flow of conversation so as to minimize the intrusiveness. The typical slow pace in general is helpful given that many elderly people experience hearing and cognitive decline along with social isolation and mistrust. To strengthen the etic approach, the interviewers could have a reflective and critical discussion about each interviewer’s beliefs, attitudes, and bias towards certain assumptions and concepts behind the interview protocol, and about how one’s urge to verify a hidden assumption can benefit from the openness towards learning of new perspectives.

Additionally, an interview also involves a set of issues related to social relationship between the interviewee and the interviewer, ranging from miscommunication, mistrust, mismatch of conceptual framework, discrepancy in values, and cultural norms of living standards, expectations, and disclosure habits (Frey and Oishi 1995; Hammersley and Gomm 2008). Sometimes, repeated follow-up interviews with the same person are needed to improve rapport, trust, and level of comfort for disclosure as well as accuracy and depth of information. Meanwhile, using a different informant (e.g., a different relative of the same family) may lead to a significant improvement in quality and/or quantity of the information acquisition, especially if the informant is cognitively or emotionally more equipped for disclosure of information. Or, sometimes instead of familiarizing the interviewee with an outsider interviewer, it might be more effective to train an insider (e.g., a village leader, or a community health worker) to be an interviewer.

Since the success of aging policy depends very much on how social systems are responsive to needs of elderly people and their caregivers, it is critical to develop reliable methods for assessment of needs. The interview is perhaps the most common approach to obtaining needs information. Given the complexity of the interview process, extra caution and care should be taken to improve the interview method in order to ascertain the quality of the data collected and our understanding of the psychological and social reality of both the interviewee and the interviewer.