Satisfaction with care among residents and families in a long-term and complex continuing care organization

Abstract

This article looks at one organization’s implementation of a satisfaction survey for all of its residents. It also explores how satisfaction ratings vary between complex continuing care (CCC) and long-term care (LTC) residents and between residents with and without dementia.

Satisfaction with care at St. Joseph’s Health Centre (SJHC) Guelph was measured during the fall and winter of 2004/2005. A total of 107 residents and patients from both LTC and CCC were interviewed and 141 family members returned mailed surveys. Overall quality of care was rated highly by both residents and families.

There were no statistically signifi cant differences comparing LTC and CCC residents and only one comparing those with and without dementia. Within the area of medical care, 57.7% of residents with dementia compared to 78.0% of residents without dementia agreed they received therapy when needed (p = .03).

This information has been very valuable to SJHC staff and administrators as they strive to improve their quality of care. The project also highlights the important work that can be achieved through collaborations between researchers and health care providers.

Introduction

Increasingly, Canadians and their government representatives are concerned with the quality of the health care system and the concept of accountability has come to the forefront of these discussions. In Ontario in 1998, the Ontario Hospital Association (OHA), in collaboration with researchers from the University of Toronto, launched the first report card reviewing acute care hospitals.1 In 2001, the first report card for complex continuing care (CCC) was published.2 A report card for long-term care (LTC) in Ontario has not yet been developed.

These reports focus on four main areas or quadrants: clinical utilization and outcomes, patient perceptions, financial performance, and system integration and change. Patient perceptions are measured using the Resident and Family Long-Term Care Survey that was developed through collaboration with researchers from Sunnybrook Health Sciences Centre and the University of Toronto as well as administrators of LTC homes.3

St. Joseph’s Health Centre in Guelph (SJHC) is a 235-bed specialty hospital and LTC centre that also offers both inpatient and outpatient CCC and rehabilitation, and several outreach programs in the community. In the fall of 2003, SJHC relocated into a new facility, and subsequently opted out of the OHA Report Card Project for CCC in that year. However, the senior administration decided that they would move ahead with an internal survey of resident and family satisfaction. This paper examines one organization’s approach in implementing a satisfaction survey for all of its residents as well as including an examination of how satisfaction ratings vary between those in LTC and CCC. A unique look at how ratings vary between residents with and without dementia is also explored.

Methods

Setting

St. Joseph’s Health Centre in Guelph is a non-profit organization with 63 CCC beds, 27 rehabilitation beds and 143 LTC beds. The facility is one of six affiliates of a healthcare system. An internal research team, from the St. Joseph’s Health System Research Network, worked with the senior administration of the facility and other staff to collect, summarize, interpret and disseminate the findings.

Survey instrument

The survey used in this study is the same as that used by the OHA in its Report Card project for CCC4. The Resident Questionnaire included 61 questions divided into seven areas: living environment (9 questions), food (9), activities (10), staff (9), dignity (9), autonomy (10), medical care and treatment (5). The questionnaire also included two overall questions asking residents and patients whether they would recommend the facility to family and friends and to rate of the quality of care and services. All questions, except the overall questions, required a “Yes”, “Sometimes”, or “No” response. A response board with a happy face, neutral face, and sad face was available to assist residents with cognitive impairment.

The Family Questionnaire consisted of 29 questions divided into four areas: living environment (6), communication with staff (8), quality of care and services (12) and resident care (3). Seven overall questions (e.g., “How would you rate the facility at taking care of your family member’s needs?”) and two questions about their background (e.g., relationship to family member) were asked at the end of the survey. The majority of questions required an “Excellent”, “Very Good”, “Good”, “Fair”, or “Poor” response.

Data collection

All residents from LTC (n= 143) and CCC (n= 63) were eligible to participate in the study and were invited to take part in an in-person interview. Residents with cognitive and/or physical limitations were not excluded from participation. Signed consent was obtained from those residents who could provide their own consent at the time of the interview. Proxy consent was obtained from family members for residents who were identified by nursing staff as being unable to provide their own consent.

The research team followed the interview and data collection procedures used by Norton et al.3 Five paid interviewers received one day of training on how to conduct interviews with residents. Topics covered during training included privacy and confidentiality, communication techniques for working with adults with hearing, vision, or speech/language difficulties, and effective ways to communicate with individuals with dementia.

Interviewers approached each resident up to three times in order to obtain an interview. These attempts were made on different days and at different times during the day. If a resident became agitated or upset, or indicated that they were becoming fatigued, the interview was stopped and the interviewer arranged to come back to complete the survey at another time. If, after three attempts, a resident was unable to respond to the questions, was too ill, or refused to be interviewed, he/she was excluded. This procedure helped to ensure that every opportunity was given to each resident to provide feedback on the quality of care experienced.

In addition, 206 surveys were mailed to family members identified by the facility as the main contact for each resident to gather their satisfaction ratings. A pre-addressed postage paid envelope was provided with the survey. Two weeks after the initial mailing, a reminder call was made by a research assistant to non-respondents.

Analyses

In the case of interviews with missing information, the data were kept for analysis if more than 50% of the questions for any domain were answered. The percentage of positive responses for a given domain was calculated by counting up the number of positive responses across all questions within a domain and then dividing by the total number of responses within that domain. The top strengths were identified by sorting percent of “Yes” or positive responses to all positively worded questions and also by identifying the percent of “No” responses to questions that were negatively worded. Areas for improvement were similarly identified by looking at those questions with the highest percent of negative responses.

All analyses were conducted using SPSS software (version 11.5). The chi-square test or Fisher’s exact test, as appropriate, were used to test for differences between LTC and CCC residents and also when comparing residents with dementia versus those without. An alpha level of 0.05 was used in all analyses.

The research protocol was reviewed and given full clearance by the Research Ethics Board of St. Joseph’s Healthcare, Hamilton.

Results

Overall, 51.9% (n = 107) of residents, including 47.5% (n = 68) of LTC residents and 61.9% (n = 39) of CCC residents, and 71.2% (n = 141) of family members took part in this project. Among individuals who could consent for themselves, the participation rate was 80.9%. In addition, 48 interviews were completed among residents with dementia, with the consent of their families. Very few partial interviews (n = 9) were deleted based on interviewer observations that indicated a lack of understanding or confusion on the part of the resident.

Resident satisfaction with care

The living environment domain had the greatest percentage of positive responses (87.1%), followed by medical care and treatment (84.8%). The activities domain had the fewest positive responses (72.1%) among residents (Table 1).

The percentage of positive responses for each domain indicated that the level of satisfaction of LTC residents was very similar to that of CCC residents (Table 1). None of the observed differences reached statistical significance; however, a trend was observed for the food domain. LTC residents were more satisfied with the food than CCC residents (81.2% compared to 70.6%, respectively). A more detailed examination of the food domain questions demonstrated that LTC residents consistently rated 5 out of 9 food questions higher than CCC residents. Two of these differences approached statistical significance:

1. taste – 20.4% more positive responses (p = .06); and

2. enough variety – 19.7% more positive responses (p = .06). It is likely that with greater sample sizes, these p-values would reach a statistically significant level.

On closer inspection of the living environment domain, LTC residents rated their satisfaction with rooms 11.9% more positively than CCC residents (94.0% “Yes” and 82.1% “Yes”, respectively), but this difference was not statistically significant.

Comparison of satisfaction of the residents with and without dementia was conducted during this study as well. Few differences were observed between the two groups (Table 2).Within the food domain, individuals with dementia tended to rate most food items higher than those without dementia. For example, 97.1% of residents with dementia as compared to 86.8% of those without dementia agreed that they are getting right amount of food, and 73.5% versus 50.7% of residents (respectively) positively responded that they are getting the types of foods they like. Within the area of medical care and treatment, the difference in satisfaction between the residents with and without dementia reached statistical significance (p= .03) for one question “Do you receive therapy if you need it?” (57.7% of residents with dementia compared to 78.0% of residents without dementia agreed they received therapy when needed).

Family satisfaction with care

The majority of family members were satisfied with the facility layout (87.6%), the resident rooms (86.9%) and agreed that SJHC was a comfortable place to visit their loved ones (84.8%). A total of 68% of family members felt that there were not enough staff and 38.5% felt that there were not enough activities for residents. Overall, roughly 80% of respondents felt that the SJHC staff were providing good, very good, or excellent care to their family members. When rating overall quality of care provided, 82% gave SJHC a rating of good, very good, or excellent.

Discussion

This project shows the value in bringing together health providers and administrators, as well as academic researchers, when assessing the quality of care. This collaboration enabled SJHC to take part in an important project to evaluate the quality of care and allowed the research to be focused on an area that was important to the facility. Collaborations of this nature can benefit the organization by providing timely information that is useful to them in targeting areas for continuous quality improvement.

The participation rate in this project was very high. Among residents who could consent for themselves, the rate was 81% and among families, the response rate was 71%. Given that both response rates were very high, this increases the likelihood that these findings are representative of all residents and families.

Overall, residents at SJHC agreed that they are receiving high quality care. On most indicators in the survey, over 70% of individuals responded that the indicator was met. Issues such as residents’ satisfaction with the food and families’ concerns about the level of staffing are concerns that are known within the LTC industry and ones that can be difficult to address.2

An important finding in this study is the fact that many residents with dementia were able to complete the interview and provide enough information for the surveys to be included in the analysis. Although few differences were noted between those with and with out dementia, having this information enabled the team to compare their responses in a meaningful way. This is an encouraging result and lends support to being inclusive when carrying out activities of this nature that address satisfaction with the quality of care.

Although patient perception is an important component of quality, satisfaction surveys have some limitations. For example, social desirability bias has been documented in the survey literature and can be especially problematic in the health care sector. Satisfaction surveys may also suffer from differential non-response bias. For example, people too ill to respond may have very different impressions about the quality of their care. Respondents tend to be over-represented by patients who had the best experience and who are functioning at the highest level.5 In general, surveys tend to overestimate the level of client satisfaction and very high levels of reported satisfaction may not truly refl ect client perceptions of their care. Given the high response rates in this study, non-response bias does not appear to have signifi cantly affected the results.

Researchers exploring client satisfaction should investigate the association between client outcomes, processes of care, and levels of satisfaction in order to reveal processes of care that lead to greater levels of client satisfaction. This was not possible in the current study, but points to an important area for future research of this type.