Israeli lay persons’ views on priority‐setting criteria for Alzheimer’s disease

Abstract

Aim  The aim of this study was to assess Israeli lay persons’ preferences for setting priorities for the care of Alzheimer’s disease (AD).

Background  As the knowledge about AD accumulates, and especially as more therapeutic and non‐therapeutic interventions for its early diagnosis and treatment are developed, health‐care costs associated with the disease rise dramatically. Therefore, setting priorities for funding these therapies, as well as other costs associated with AD is becoming an increasingly complex need.

Methods  A convenience sample of 624 community‐dwelling men and women participated in the study (mean age = 49, range = 20–90). Participants were asked how strongly they agreed or disagreed with 18 criteria that would be used to set priorities for the allocation of resources in the area of AD.

Results  High‐weight criteria included being a life‐threatening condition and the benefit of treatment. Average‐weight criteria included the severity of the disease, treatment costs and equity of access. The age of the patient also was highly rated. All other patient‐related criteria were rated as low. Value orientations and education were the main variables associated with participants’ preferences.

Conclusions  The lay public seems to endorse a multi‐criteria decision process for the allocation of resources in the area of AD. Similar to other diseases – disease‐related criteria were highly preferred. These preferences should be compared with those of other stakeholders such as clinicians and policy makers.

Introduction

Alzheimer’s disease (AD) is the most common cause of dementia ¹ and is associated with significant health‐care costs. Cost‐of‐illness studies conducted in different countries have shown that these costs are largely for institutionalization and for assistance with the performance of activities of daily living in the community. ² , ³ , ⁴ , ⁵ , ⁶ Additionally, the costs of the disease increase with functional and cognitive deterioration and are especially high for the management of behavioural and psychological symptoms associated with the disease. ⁵ , ⁷

As the knowledge about AD accumulates, and especially as more therapeutic and non‐therapeutic interventions for its early diagnosis and treatment are developed, ⁸ the health‐care costs associated with the disease rise dramatically. Therefore, setting priorities for funding these therapies, as well as other costs associated with AD, is becoming an increasingly difficult challenge for health‐care systems worldwide. ⁹ Given the complex decisions facing health‐care systems in the area of AD, an important question is what are lay persons’ preferences regarding the criteria that should be used to guide priority‐setting decisions?

Before presenting the study methods and results, we first turn attention to three areas of literature relevant to this study: priority setting in health care and the role of lay persons’ preferences in this process, the need for priority setting in the area of AD and the experience of Israel in priority setting in health care.

Priority setting in health care

Setting priorities in health care is not a new issue, but its importance is growing as a result of technological developments, increasing health needs, escalating costs and constrained resources. ¹⁰ In this situation, policy makers are confronted with difficult decisions related to which diseases, technologies and populations will be given priority in the use of public funds.

Although priority‐setting decisions are associated with competing values, there is a consensus that any acceptable and successful priority‐setting process should be based on legitimacy and fairness. ¹¹ To assure these fundamental characteristics, the need to include a wide range of stakeholders, and not only technical experts, is strongly recommended. ¹² Consequently, in most countries, although lay persons do not have a direct and formal impact on priority‐setting decisions, their views and values are now being considered in an attempt to enhance the legitimacy and fairness of the decision‐making process while reinforcing trust and confidence in the health‐care system. ¹³ The main factors influencing lay persons’ preferences for resource allocation were found to be socio‐demographic characteristics (such as age and gender) and their moral values. ¹⁴

Alzheimer’s disease and the need for priority setting

Neurodegenerative dementing disorders in general and AD (the most common type of dementia), in particular, have been lately defined to be an increasing public health epidemic. ¹⁵ , ¹⁶ It is estimated that approximately 24 million people are affected by a dementing disorder worldwide, and the number is projected to double every 20 years. ¹⁷ These numbers are accompanied by enormous economic costs to individuals and to society, with recent estimates reporting the worldwide direct costs of dementia care to be over $300 billion. ¹⁸

With these escalating expenditures, it is clear that health systems worldwide are confronted with the need to make rational and appropriate resource allocation decisions. Indeed, in some countries, such decisions are already being taken. For example, the UK’s National Institute for Health and Clinical Excellence (NICE) has recently decided to restrict access to drugs for AD to people in the moderate stages of the disease. Although this decision was strongly and militantly opposed by several experts and patient groups, NICE’s judgement has more recently been upheld in court. ¹⁹

This example stresses the importance to expand the knowledge about rationing health care in general and the opinions of lay persons in particular in the area of AD.

The experience of Israel in priority setting

In 1995, Israel enacted the National Health Insurance (NHI) law, which brought a reform to the Israeli health‐care system. The main goals of the new law were to increase equity among different populations and to improve services for weaker populations. Additional aims of the law were ‘…to make universal and to clarify the rights of citizens to health insurance coverage, to increase citizens’ freedom of choice among sick funds…, to stabilize the system financially, to improve the quality of care, and to free the Ministry of Health from operational responsibilities so that it could devote more efforts to monitoring the system,’ ²⁰ (pp. 56–57).

Under the NHI law, all the sick funds (similar to the Health Maintenance Organizations in the USA) are required to provide the same basic services (called basket of services) to their members. Services not included in this basket can be obtained at an additional cost. Although at the time it was enacted, the NHI did not explicitly define a process to make decisions about the services to be allocated in the basket of services; today, a National Advisory Committee (including officials from the Ministry of Health, the Ministry of Finances, the sick funds and public representatives) makes these decisions based on considerations such as the potential of the technology or service to prevent mortality or morbidity, the number of patients to benefit, the financial burden on society and the individual patient. ⁹

In sum, given the importance of priority setting in general and the scant theoretical and empirical attention given to the study of prioritization in the area of AD in particular, the aim of the present study was to examine Israeli lay persons’ preferences regarding the allocation of resources in the case of AD.

Methods

Selection of method

Several methods have been described and used for eliciting lay persons’ preferences for setting priorities in health care. These include self‐completion questionnaires or interviews in surveys with large samples as well as focus groups and other qualitative approaches such as in‐depth interviews with small samples. ¹⁴ As the aim of this study was to obtain results generalizable to the target population rather than to ‘set an agenda’, ¹⁴ we used a survey method as the most useful method to explore what members of the Israeli lay public think about the topic while allowing a wide cross‐section of the general public to be approached.

Participants

A convenience sample of community‐dwelling men and women aged 20 years and over participated in the study. All participants were unpaid volunteers who were approached by five trained research assistants in three large cities in the northern part of Israel. The research assistants were each instructed to recruit 150 participants. Overall, 624 adults consented to participate in the study, rendering a response rate of 83.2%. Reasons for non‐participation included lack of interest and time constraints.

The majority of the participants were women, born in Israel, married and had at least 11 years of education. The socio‐demographic details of the participants are shown in Table 1, together with 2007 census data for the Israeli population. ²¹

Table 1.

 Distribution (%) of the main socio‐demographic characteristics in the study’s sample and in the Israeli population aged 15 years and over

Characteristic	Sample (n = 624)	Israeli population aged 15 years and over1
Gender
Male	48.2	48.6
Female	50.9	51.4
Age
20–34	26.4	37.7
35–54	31.7	35.2
55–74	33.4	33.4
75+	7.2	8.5
Number of years of education
0	2.1	2.1
1–4	1.1	1.0
5–8	7.0	6.0
9–10	5.7	9.4
11–12	22.8	35.3
13–15	25.3	24.3
16+	35.9	21.9
Place of birth
Israel	68.9	58.6
Asia/Africa	7.9	13.1
Europe/America	22.0	28.3
Marital status
Not married	29.5	34.3
Married	71.5	65.7

¹From the Central Bureau of Statistics. ²¹

Although random sampling is a desirable technique for minimizing bias, it was not feasible in this study because of financial constraints. However, as can be observed in Table 1, the study sample was very similar to the Israeli population, although there was a lower percentage of persons aged 20–34 years, a higher percentage of persons with at least 16 years of education and a lower proportion of non‐married participants.

Measures

A structured questionnaire was developed for the study (the questionnaire is available from the author at request) to include the most relevant factors associated with priority setting. The following measures were included.

Criteria for priority setting

Based on the literature on priority setting, ²² participants were asked to rate to what extent they considered each one of the items in a list of 12 reasons for the allocation of health resources to be a criterion for priority setting of resources in AD (Table 2). The criteria included patient‐related criteria (patient’s age, socio‐economic status, gender, religion, power of influence and responsibility for causing the disease), disease‐related criteria (severity of the disease, being a life‐threatening condition, cost of the treatment and its benefit) and society‐related criteria (equity of access and political views). Participants were asked to rate each criterion on a 5‐point scale, ranging from 1 = definitely not a criterion to 5 = definitely a criterion for setting priorities of resource allocation to AD.

Table 2.

 Respondents’ rating of priority‐setting criteria for AD (n = 624)

To what extent do you consider each one of the following to be a criterion for setting priorities in the allocation of resources for AD?	1	2	3	4	5	No response
	Definitely not a criterion				Definitely a criterion
Patient‐related criteria
Age	24.2	6.9	18.4	18.1	30.6	1.8
Socio‐economic status	40.7	10.9	14.3	14.7	17.8	1.6
Gender	81.7	3.5	7.1	4.0	1.8	1.9
Religion	90.5	2.1	2.9	2.2	1.1	1.1
Power and influence	64.3	10.6	10.9	6.7	6.3	1.3
Responsibility for causing illness	52.2	13.3	14.3	12.3	6.4	1.4
Disease‐related criteria
Severity of the disease	13.0	6.5	17.5	26.9	34.6	1.4
Life‐threatening condition	10.3	5.8	12.5	19.4	51.0	1.0
Treatment costs	26.9	14.6	21.3	16.5	18.9	1.8
Benefit of treatment	7.9	7.4	15.9	24.7	42.1	2.1
Society‐related criteria
Equity of access	23.7	8.0	13.6	16.2	34.1	4.3
Political views	88.6	3.8	2.9	1.1	1.3	2.2

Participants’ characteristics

Information was collected about participants’ value orientations, perceived threat about developing AD and socio‐demographic characteristics.

Value orientation:  As the literature on priority setting clearly stresses the importance of values to this process, ²³ participants’ personal values were assessed using a list of 18 behavioural patterns, validated by Schomerus and colleagues, ²⁴ and representing three different value orientations: liberal values, traditional values and modern values. Participants were asked to rate the desirability of each one of the values using a 7‐point scale, ranging from 1 = not desirable at all to 7 = very much desirable.

Following a principal components factor analysis of the 18 items, and similar to Schomerus and colleagues, three indices were derived by calculating the means of the items in each index. The first index reflected liberal values (such as being just, treating everyone equally, removing social differences between individuals, being conscientious, being tolerant, being emotional and solving conflicts by compromise) and had excellent internal reliability (Cronbach’s alpha = 0.84). The second index reflected traditional values (such as accomplishing one’s tasks, being on time, being careful, trying one’s best and valuing wealth) and had very good internal reliability (Cronbach’s alpha = 0.76). The third index reflected modern values (such as enjoying life, being independent, doing as you please and seeking self‐realization) and had good internal reliability (Cronbach’s alpha = 0.74).

Perceived threat

As the threat posed by the disease is reported as one of the factors affecting priority setting, ²² participants’ perceptions regarding the likelihood of developing AD were assessed using two items: (1) Perceived likelihood was assessed by asking participants how likely it was that they would develop AD at some point in their lifetime. (2) Perceived concern was assessed by asking participants how much they worried about developing AD at some point in their lifetime. Each item was rated on a 5‐point Likert‐type scale, with higher scores indicating perceptions of greater threat. ²⁵ , ²⁶

Socio‐demographic characteristics:  These included gender (female/male), age and years of education.

Procedure

Face‐to‐face interviews were conducted using a structured interview. Although the study was not approved by a research committee, similar to other studies ²⁷ informed consent was considered to have been obtained when potential participants agreed to answer the questionnaire and the survey procedure adhered to the highest research standards and maintained the highest ethical principles in its procedures and methods. After receiving a thorough training, the following steps were followed by the interviewers: they introduced themselves, briefly described the survey topic, identified the person and organization conducting the research, described the purpose of the research and gave a ‘good faith’ estimate of the time required to complete the interview, they also promised anonymity and confidentiality, mentioned to the participant that participation was voluntary and that item non‐response was acceptable. Finally, after asking verbal permission, they began with the interview.

Statistical analyses

Data were analysed using spss statistical software (SPSS Inc., Chicago, IL, USA). Descriptive statistics (means, SD and percentages) were used to describe the participants and the main variables.

Following ²² three cut‐off points were defined in order to characterize the relative importance of each priority‐setting criterion presented to the participants, as follows:

• • A criterion was defined as a low‐weight criterion, if 0–33% of the participants accorded a score of 4 or 5 to the criterion.
• • A criterion was defined as an average‐weight criterion, if 34–66% of the participants accorded a score of 4 or 5 to the criterion.
• • A criterion was defined as a high‐weight criterion, if 67–100% of the participants accorded a score of 4 or 5 to the criterion.

Next, the 5‐point scales were dichotomized, with responses ranging from 1 to 3 coded as 0 (not a criterion) and responses ranging from 4 to 5 coded as 1 (definitively a criterion). Associations between the agreement with each one of the items in the three types of criteria (e.g. equity of access and political views in the society‐related criteria) and participants’ characteristics (i.e. age, gender, education, the three scales in value orientation and perceived threat) were examined using logistic regressions.

Results

Criteria for the prioritization of resources in AD

The percentages of participants not providing an answer to the specific criteria were relatively low (ranging from 1 to 4%; Table 2). The majority of the participants strongly agreed that disease‐related aspects were definitive criteria to be used for the prioritization of resources in AD. Among these criteria, being a life‐threatening condition, severity of the disease and benefit of the treatment were the most important criteria endorsed by the participants (70, 62 and 67%, respectively reflecting a score of at least 4 for each criterion). There was also a general agreement regarding patient‐related criteria. With the exception of age, the majority of the participants agreed that these factors should not affect decisions regarding priority setting in the area of AD. Similarly, as many as 90% of the participants did not consider political values to be a factor that should affect priority‐setting decisions. There was wide variation in the distribution of responses regarding other criteria.

Table 3 shows that disease‐related criteria were considered the most important criteria, with one‐third to two‐thirds of the participants rating these criteria as having high or close to high weight. Patient‐related criteria were accorded low to average weight for priority setting, with age being the most important criterion in this category. Among society‐related criteria, equity of access was accorded moderate weight, whereas political values were accorded low weight.

Table 3.

 Weighting of criteria for priority setting according to the number of respondents in agreement

	Patient‐related criteria	Disease‐related criteria	Society‐related criteria
High weight (>66% agreed)		Life‐threatening condition (71%)
		Benefit of treatment (68%)
Average weight (33–66%agreed)	Age (49%)	Severity of the disease (62%)	Equity of access (53%)
		Treatment costs (36%)
Low weight (<33% agreed)	Gender (6%)		Political views (2%)
	Socio‐economic status (32%)
	Power and influence (13%)
	Responsibility (19%)
	Religion (3%)

Factors associated with participants’ preferences for criteria for the prioritization of resources in AD

Regression analyses revealed a few, but interesting, associations (Table 4). Patient‐related criteria were mainly associated with value orientations. Participants reporting higher liberal values considered patient‐related factors to be less important criteria for setting priorities in AD. Higher levels of modern values were associated with attributing higher importance to personal responsibility for the disease as a priority‐setting criterion. Finally, persons who expressed a higher concern about developing AD attributed less importance to religious background as a criterion for resource allocation in AD as indicated by a statistically significant odds ratio of 0.3.

Table 4.

 Logistic regressions showing the association between participants’ characteristics and their agreement with the importance of each criterion for priority setting

Criteria	Participants’ characteristics
	Education	Liberal views	Modern views	Perceived concern about developing AD
Patient‐related criteria1
Religion	–2	0.5* (0.2–0.8)3	–	0.3** (0.1–0.9)
Power and influence	–	0.5* (0.4–0.7)	–	–
Responsibility for causing illness	–	0.7** (0.5–0.9)	1.5* (1.1–2.0)	–
Disease‐related criteria4
Life‐threatening condition	–	0.7** (0.5–0.9)	1.4* (1.1–1.7)	–
Treatment costs	–	–	1.3** (1.0–1.6)	–
Benefit of treatment	–	–	–	–
Society‐related criteria
Equity of access	1.0** (0.9–1.0)	–	–	0.6** (0.4–0.9)
Political views	0.8** (0.7–0.9)	–	–	–

¹No statistically significant associations were found with age, socio‐economic status and gender.

²Cells with dashes indicate statistically non‐significant results.

³Numbers indicate odds ratios (95% CI).

⁴No statistically significant associations were found with severity of the disease.

*Significant at P < 0.01; **Significant at P < 0.05.

Value orientations were also the main factors associated with disease‐related criteria. Higher levels of liberal values were associated with attributing lower importance to the severity of the disease, whereas higher levels of modern values were associated with attributing higher importance to this criterion as well as to the cost of treatment for the disease.

Interestingly, society‐related criteria were not associated with value orientations, but rather with participants’ level of education and with their concerns about developing AD. Higher levels of education were associated with attributing increased importance to equity of access and attributing less importance to political views. Finally, higher concerns about developing AD were associated with attributing less importance to equity of access.

Discussion

Although the lay public is increasingly being considered as one of the main stakeholders in the process of setting health‐care priorities, ²⁸ there is a dearth of studies assessing their views in the area of AD. The aim of the present study was to address this gap in the literature. The findings showed that based on the small percentage of refusals to participate, lay persons can and want to express their preferences for the criteria to be used in guiding priority‐setting decisions for AD.

Participants attributed average to high importance to a wide variety of criteria for priority setting. This finding supports the need stressed in recent years for a multi‐criteria decision‐making process that will allow policy makers to make decisions while considering a broad range of criteria simultaneously. ¹⁰ In so doing, lay persons’ understanding of the complexity of setting health‐related priorities in general and in AD in particular may be recognized.

Similar to other studies examining public views on priority setting, ²² , ²⁹ , ³⁰ disease‐related criteria were accorded high weight by the participants. Among these criteria, being a life‐threatening condition, severity of the disease, and benefit of the treatment were considered the most important factors in the allocation of resources for AD. These factors might reflect the desire of the participants to take into account, first and foremost, patients’ quality of life in the decision‐making process. Thus, the main criteria guiding lay persons’ preferences for priority setting in the area of AD are those associated with the characteristics of the disease.

Treatment costs were endorsed as a criterion for priority setting in the area of AD by only a third of the participants. This percentage is much lower than that reported by other studies. ²² Several explanations can be offered for this discrepancy. The first explanation relates to the lay public’s overall preferences for prioritization in health care. As reported in the literature, ³¹ the public is frequently guided by a desire to give equal priority to all members of the population, and as such might attribute little importance to the cost of treatment. This explanation is reinforced by the finding that more than 50% of the participants in this study reported equity of access as one of the criteria for priority setting.

The second explanation relates to the characteristics of the Israeli health‐care system. As stated in the ‘Introduction’ section, Israel’s NHI law assures health insurance coverage and therefore enhanced equality to all citizens. ¹⁹ Under such a system, members of the lay public might be less preoccupied by the cost of treatment as a factor affecting the allocation of resources.

The last explanation relates to the distinctiveness of AD. During recent years, we have witnessed a notable increase in scientific and clinical knowledge in the treatment of AD. Progress has been made in the development of both pharmacological and non‐pharmacological approaches. ³² However, these treatments do not cure the disease. As it has been shown that lay persons have a fair knowledge about AD in general and about its treatment in particular, ³³ , ³⁴ it might be assumed that the low importance attributed to treatment costs is associated with this knowledge.

Almost half of the participants considered the patient’s age to be a criterion for priority setting in the area of AD. Age has been extensively discussed as a debatable factor for setting priorities, ²² but our results add little to the body of knowledge in this area as we did not examine which age should receive higher priority. Future studies should clarify whether the public’s views about age are determined more by the need to assure quality of life to young persons with early onset dementia or to cope with old persons in the most severe stages of the disease. Given the egalitarian approach of the Israeli NHI law, it is not surprising that besides age, other patient‐related criteria were accorded low weight.

Several interesting associations were found between participants’ characteristics and their attributions of importance for each criterion. The main factors associated with the participants’ views about the different criteria were their value orientations. Not surprisingly, participants reporting higher liberal values endorsed less patient‐related criteria. The items used to assess liberal values in this study reflect the principles guiding liberal egalitarian theories of distributive justice. These theories argue the need to eliminate inequalities that arise from factors outside the individual’s control, such as socio‐demographic and genetic characteristics. ³⁵ Therefore, it is understandable that persons endorsing these values will attribute less importance to their impact on resource allocation. A similar explanation might be applied to the association between liberal values and severity of the disease.

Modern values, reflecting independence, self‐realization and the value of individual choice, were associated with the attribution of higher importance to personal responsibility for the disease as a priority‐setting criterion. The association between higher moral values and agreement with disease‐related criteria is less obvious, although increasing attention is being devoted to the influence of life style and health behaviour on the prevention of AD. ³⁶ This might explain the finding that persons attributing higher value to independence and choice are more inclined to agree with using severity of the disease as a criterion for priority setting, if they think that the severity of AD is associated with personal choices regarding life style. As for the association with treatment costs, the value placed on wealth as part of modern value orientations may provide an explanation.

A modest association was found between higher education and agreement with equity of access as a criterion, whereas lower education was associated with attributing more importance to political views. These associations might be a reflection of the relationship between education and value orientations. ³⁷ Finally, increased concern about developing AD was associated with less agreement with religion and equity of access as criteria. This might reflect the desire of those concerned with developing the disease to secure the allocation of resources even at the cost of reducing egalitarian principles.

Limitations

This study has several limitations. Firstly, the sample was not randomly selected and thus may have been subject to selection bias. Indeed, the sample was better educated and more likely to include married participants than the Israeli population. No information was gathered regarding the participants’ income status and economic activity and no information is available about the characteristics of non‐respondents. Secondly, the survey method used did not allow respondents to explain their views and expand on their preferences. ³⁸ This is especially critical given the importance and complexity of the topic. Moreover, the criteria presented to the participants (such as ‘power of influence’ and ‘benefit of the treatment’) were not explained to them, so that the results reflect the understanding of each respondent of these terms. Future studies using other methodologies, such as focus groups or a combination of qualitative and quantitative methods, may more effectively capture the profound nature of these issues. Finally, although participants were specifically asked regarding their preferences for prioritization in AD, they were not asked specifically about prioritization of different treatments, diagnostic criteria, etc.

Despite these limitations, the present study addressed a complex and understudied topic – the public’s preferences for the criteria guiding resource allocation in AD. In light of the demographic changes forecast for the future, increasing numbers of persons worldwide can be expected to develop AD and other dementias. Policy makers are already confronted with difficult decisions when trying to establish a legitimate and fair process of priority setting for the allocation of health resources in AD. In view of the controversies and anger that have accompanied NICE’s decisions about the prioritization of drugs for AD, ¹⁹ it is obvious that research about lay persons’ preferences about this process should be encouraged and expanded. They are important for procedural justice, and may help to both legitimize the process and resulting priorities and expand accountability.