以功能狀況改變作為長期照護之品質指標的分析

行政院國家科學委員會補助專題研究計畫成果報告

※※※※※※※※※※※※※※※※※※※※※※※※※

※ ※

※ 以功能狀況改變作為長期照護之品質指標的分析 ※

※

※

※※※※※※※※※※※※※※※※※※※※※※※※※

計畫類別：ˇ個別型計畫 □整合型計畫

計畫編號：NSC89－2314－B－110－005－

執行期間：89 年 08 月 01 日至 92 年 1 月 31 日

計畫主持人：葉淑娟

共同主持人：

計畫參與人員：楊長興、林妍如、蔡淑鳳

本成果報告包括以下應繳交之附件：

□赴國外出差或研習心得報告一份

□赴大陸地區出差或研習心得報告一份

□出席國際學術會議心得報告及發表之論文各一份

□國際合作研究計畫國外研究報告書一份

執行單位：國立中山大學人力資源管理研究所

中文摘要 本研究主要探討護理之家住民的特質、機構的特質、與住民的功能狀況改變的關係。 本研究共涵蓋 311 高雄市的有效樣本。我們發現 194 位住民的 ADL 並未有改變，54 位 身體功能變好，60 位身體功能變差。機構的大小由 10 床到 168 床不等。調查的住民之 平均年齡為 72.14 歲，平均住機構的時間為 42 天。多變量的分析得知：年輕的和身體功 能依賴項目較少的住民較有機會改善其身體功能﹔ADL 初評分數較高、從未轉入急性醫 院的住民均有較高的機會可以維持住院前的身體功能甚至改善原有的功能。另外，機構 的大小明顯的與品質的好壞有關。因而本研究也認為國人仍然明顯的相信大的機構之照 護品質會較好的觀念。 Abstr act

This study investigated the association between personal characteristics, facility characteristics, source of admission, and functional change of elderly nursing residents. This study is a longitudinal study with nursing home resident as the unit of analysis. We first interviewed 454 nursing residents from 35 nursing homes on June of 2001 and then followed the same 454 cases to December of 2001. We lost 60 cases in the second interview due to death and 83 cases because of discharge. It made the final study sample containing 311 nursing home residents. The outcome variable is ADL change, which was defined as the difference between the second ADL score and the first ADL score. The ADL change was categorized into three levels: functional improvement, stablization, and functional decline. We used multinomial logistic regression to examine the odds of ADL change as a function of the covariates using SPSS.

We found that 194 nursing residents maintain no change in ADL scores, while 54 and 60 showed functional improvement and functional decline, respectively. Facility size ranged from 10 to 168 beds. The average age of the sample was 72.14 years with length of stay 42 days, and most of their length of stay around 42 days. When comparing with functional decline (means that functional status is worse), the younger residents (Odds Ratio (OR) = 0.96, 95% confidence interval (CI) = .94-.99) and the less items on dependence (OR=.41; 95% CI=.26 -.67) are related to functional improvement. Residents with the higher ADL score of first time assessment (OR=1.46; 95% CI=1.22-1.74 for stablization; (OR=1.35 and (95% CI=1.13-1.62 for negative change, respectively) and never transferring to acute care hospitals (OR=2.56; 95% CI=1.26-5.19 for stablization; OR=2.57; 95% CI=1.21-5.46 for functional improvement, respectively) are more likely to maintain or improve their functional status. Comparing with functional decline, we found that the larger the facility size (OR=1.014; 95%CI =1.001-1.027), the better the residents’ functional status is.

Based on the results of this study we conclude that residents’ age, first time ADL score, having transferring to acute care hospitals, number of dependent items are important predictors of changing functional status. In addition, facility size is also a significant indicator to predict functional status. This could be culture difference because people trust the quality of care from larger institutions than from small ones in Taiwan.

Introduction

There are widespread concerns regarding the quality of nursing home care and whether care is improving. Research on predictors of change in functional status has emphasized the role of the resident’s level of functioning at the time of placement, duration of

institutionalization, and extent of cognitive impairment (Mitchell, 1978; Rudman et al., 1993 & 1994; Kane et al., 1996). However, knowledge of the relative effect of these factors on

functional status is still limited. The purposes of this study are to examine resident and facility attributes associated with functional status change, as well as to investigate the predictors of functional change.

Methods

DATA

This study was a longitudinal study with nursing home resident as the unit of analysis. The interviewers’ training involved both a general introduction to the survey interviewing and fieldwork techniques and procedures and a review of the specific aspects of the study that interviewers were responsible for. We first interviewed 454 nursing home residents from 35 nursing homes on June of 2001 by trained registered nurses. Then, we followed the same 454 cases to December of 2001 and performed the second fact-to-face interview by the same registered nurses. We lost 60 cases in the second interview due to death and 83 cases because of discharge. It made the final study sample containing 311 nursing home residents.

We collected data on demographic information, such as age, marital status, education, occupation, income, and religion. We also collected information in comorbidity, source of admission, alternative treatment, rehabilitation services, number of transferring to hospitals, and facility size. In addition, functional status, number of dependent items, and cognition function were measured. The outcome variable is change in functional status (ADL change). Change in functional status was examined using two assessment periods. We defined ADL change as the difference between the second ADL score and the first ADL score. The ADL change was categorized into three levels: functional improvement, stabilization, and functional decline.

ANALYTICAL APPROACH

We first examined differences in ADL change by demographic variables. We also examined the association between ADL change and all demographic and health-related variables. The stepwise multinomial logistic regression was used to regress ADL change on independent variables and all covariates to find the significant predictors (used α level at 0.10). Initial analyses included age, gender, education, religion, occupation, income, and

comorbidity variables as covariates. However, some of these parameters were not significant predictors and they were subsequently dropped from further analyses. Finally, a parsimonious multiple logistic regression analysis was used to determine the odds of ADL change as a function of the covariates. All the analyses used SPSS statistic software.

Results

We found that 194 nursing residents maintain stabilization in ADL scores, while 54 and 60 showed improvement and decline, respectively. Facility size ranged from 10 to 168 beds. The average age of the sample was 72.14 years, and most of their length of stay around 42

days. About 30% experienced at least one time transferring to acute care hospitals due to diseases. Most of the residents were self-referred to nursing homes, while 8.02% was recommended by hospitals and 5.12% was referred by the Dept of Social Welfare. Only 2%

residents had tried alternative treatment and 40.97% received rehabilitation services. The average of dependent items was 4.2 with the mean score of initial ADL score 10.14. Around

6.6% had pressure ulcer and less than 1% needed care on tracheotomy, nasogastric feeding, and urine catheterization. The average number of co morbidities was 0.95, range from 0-4.

Model 1 compared stabilization with function decline. We found that the younger residents (Odds Ratio (OR) = 0.97, 95% confidence interval (CI) = .94-.99) and the less items on dependence (OR = .41; 95% CI =.26 -.67) are related to functional improvement. Residents

acute care hospitals (OR=2.56, 95%CI = 1.26-5.19) were more likely to maintain their functional status.

On the other hand, comparing functional improvement with functional decline (Model 2), we found that higher initial ADL score (OR = 1.35, 95% CI =1.13-1.62) and never transferring to acute care hospitals (OR = 2.57; 95% CI = 1.21-5.46 for) had likelihood to

improve their ADL function. In addition, the larger the facility was (OR=1.014; 95% CI =1.001-1.027), the better the residents’ functional status would be. Other predictors, such as cognition function, length of stay, receiving rehabilitation services, having Chinese alternative

treatment, number of co morbidities, and sources of admission, had no statistical significant results (not shown). The two parsimonious models showed in Table 2.

Discussion

The results from this study also suggest that initial functional status can predict functional change while controlling for case-mix and other confounding factors. This is consistent with current literature review. Namely, no matter where the nursing home residents are, the initial functional status can be a good indicator to predict residents’ functional change. Never transferring to other acute care hospitals has been shown to be strongly predictive of functional improvement in our study. There will be a clear distinction between whether or not transferring to acute care setting during institutionalizing in LTC facilities. Residents without transferring to hospitals mean that she/he is increasing functional stability. On the other hand, visiting hospitals again means more medical or mental problems, which has shown evidence in influencing the risk of functional dependence.

Facility size showed positive association with functional improvement. This might be culture difference. In Taiwan, people trust the larger facilities more than smaller facilities because they assume that the large institutions can provide better quality of care based on the assumption of having more resources in terms of staff, financial, and facilities. Age is another predictor for functional change. However, we have to be caution when explaining the strength of age in predicting change in functional status. Based on our results, younger residents showed more likelihood to maintain their functional status than older residents. On the other hand, there is no evidence to indicate that the younger residents are more likely to improve their functional change.

We used 6 months as the cut off point for length of stay and found that there were no association between the length of stay and ADL change. It is quite different with previous literature because most of our study sample stayed more than 180 days in nursing homes, indicating more functional limitations. Regarding the comorbidity issue, our results did not support residents with more medical conditions would show higher risk in functional decline. Although 41 percent of our study sample had received rehabilitation services, there were no association between the functional change and the services. It should interpret the results carefully because the data did not collect detailed information on rehabilitation. Instead, we only asked “did you receive any rehabilitation service lately?” We believe that different rehabilitations, such as range of motion, will change the association.

Limitation

enhances the knowledge of functional status for Taiwan’ elderly population by investigating the predictors of changing in functional status. However, the limitations still not addressed in this study include the fact that some variables may provide only attenuated information when dealing with functional status issue. For example, number of co morbidities provided the quantity measurement; however, it did not give the specific information on which comorbidity might real impact on the functional status.

This study took a limited view of outcomes, focusing on functional changes. Estimates of the impact on other resident outcomes such as mood state, life satisfaction, and autonomy, could not be made with these data. In addition, other outcome measures such as the incidence of and recovery from pressure sores, contractures, or urinary-tract infections, would broaden the view of quality of care in nursing homes.