The Views of Health Care Professionals and Laypersons Concerning the Relevance of Health-Related Behaviors in Prioritizing Patients

Abstract

This article aims to investigate to what extent health care professionals in Portugal find health-related behaviors relevant as a criterion to priority setting, to study what type of risk behavior they consider relevant in such decisions and to compare their views with those of laypersons. An online questionnaire was used to collect data from a sample of 178 health care professionals and 295 laypersons. The statistical analysis was performed through the application of generalized linear models and logistic regressions. Health professionals consider more than laypersons that information about health-related behaviors is relevant in prioritization between individuals. Both groups regard information about illegal drug abuse, excessive alcohol consumption, and smoking more important to priority setting decisions than information about overweight or lack of physical exercise. The findings are important for the definition of rationing policies as the professionals’ decisions are those that, ultimately, influence health expenditure.

Keywords

attitudes of health professionals attitudes of laypersons health care rationing health-related behaviors personal responsibility

The costs of chronic and noncommunicable diseases (NCD) are increasingly threatening the financial sustainability of publicly funded health systems. NCDs are the result of a combination of genetic, physiological, environmental, and behavioral factors. The European region is the most affected by NCD (World Health Organization [WHO], 2017). Cardiovascular diseases account for most NCD deaths, around 17.7 million people annually, followed by cancers (8.8 million), respiratory diseases (3.9 million), and diabetes (1.6 million). These four groups of diseases account for over 80% of all premature NCD deaths. Tobacco use, physical inactivity, the harmful use of alcohol and unhealthy diets all increase the risk of dying from an NCD (WHO, 2017). In the European Union, diseases of the circulatory system and cancer were, by far, the leading causes of death in 2014. Diseases of the circulatory system include those related to high blood pressure, cholesterol, diabetes, and smoking. Heart diseases accounted for 126 deaths per 100,000 inhabitants across the EU-28 in 2014 (Eurostat, 2017).

This means that increasingly scarce health care resources are spent in diseases that to some degree could be avoided through individual health-related behavior changes. The debate on responsibility for health is flourishing in the same proportion as the scarcity of health care resources is becoming evident. The lifestyle solidarity that used to be hidden in the past is increasingly being questioned (Trappenburg, 2000). More governments are adopting interventions to change health behaviors, most of which have proved to be effective (Jepson, Harris, Platt, & Tannahill, 2010, for a review). Even so, and particularly in public health systems, the debate that prevails among policy makers has to do with the possibility of rationing health care based on individual health behavior (Trappenburg, 2000; Sassi & Hurst, 2008; Schmidt, 2009; Schmidt, Asch, & Halpern, 2012; Schmidt, Voigt, & Wikler, 2010). Holding individuals accountable for their choices in the context of health care is, however, controversial (Callahan, Koenig, & Minkler, 1999; Cappelen & Norheim, 2005; Minkler, 1999).

There is a growing empirical literature on the views of the general public concerning the use of health-related behaviors choices of people whose illness is related to those choices as a criterion to take bedside health care rationing decisions (Exel, Baker, Mason, Donaldson, & Brouwe, 2015; Gu, Lancsar, Ghijben, Butler, & Donaldson, 2015; Olsen, Richardson, Dolan, & Menzel, 2003; Pinho, Borges, & Zahariev, 2017; Rogge & Kittel, 2016; Stegeman, Willems, Dekker, & Bossuyt, 2014). The results seem to indicate that this idea is far from being consensual. There are few literature exploring systematically this question for a broad range of unhealthy behaviors (Bringedal & Feiring, 2011; Miraldo et al., 2014), and no study was found comparing and contrasting the views of health care professionals with the views of laypersons without specialist expertise in health care. This is important because, at a micro level, health care professionals are those who ultimately decide which patients should have priority. These decisions of health professionals influence the increase or decrease in health expenditure. If health care professionals hold systematically different ethical views about health care rationing than the general public and policy makers, this would be a cause for concern. This is also a matter of procedural justice, in facilitating transparency and helping to ensure that open and fair decision-making processes are followed (Daniels & Sabin, 1997). Hence, it is important to know if differences exist between those who decide and those who are affected by those decisions, so that processes can be redesigned to ensure that one set of views is not predominant.

This study will contribute to this knowledge gap by exploring the views of health professionals (those directly involved in bedside rationing decisions—physicians and nurses) in Portugal about the relevance of five different type of risk behaviors (smoking, excessive alcohol beverage consumption, overweight/obesity, illegal drug use, and lack of physical exercise) in prioritization between individuals, and compare them with the views of members of the general public without professional expertise or experience working in the health care field. It is the first time that Portuguese health professional’s attitudes about different personal responsibility issues have been assessed in a systematic way. This study is particularly relevant for Portugal for two main reasons. First, due to the external bailout to which Portugal was subjected (2011-2014), with a reduction in health spending that reached more than 5% per year in real terms (OECD, 2015). However, the health care demand has not ceased to increase, a reality that has been pressing policy makers to manage the health sector in an efficient manner, at the basic level of resource allocation. Second, Portuguese society has been adopting unhealthy behaviors. Recent data revealed that Portuguese are mostly sedentary, their daily food intake is unbalanced (rich in sodium and sugar), they have the highest alcohol consumption rates in Europe and the percentage of smokers and illegal drug abusers is around 20% and 9.5%, respectively (DGS, 2016; Lopes et al., 2017; SICAD, 2016).

Method

With this study we intend to explore and compare attitudes of Portuguese health professionals and laypersons (without medical knowledge) about three specific concerns: (1) Are health-related behavior relevant to priority decisions? (2) Which type of risk behavior respondents find most relevant to influence priority decisions? and (3) As regards the financing of the National Health Service (NHS), which type of risk behavior is more penalized? Moreover, by taking the whole sample we investigate how different views correlate with two sociodemographic characteristics of the respondents: gender and age. For the age variable, we consider 44 years old, the median age of the Portuguese population (Pordata, 2018), and compare the views of those who are younger than 44 years with those who are older than 44 years.

Questionnaire

Data were collected through a self-reported online questionnaire. The questionnaire uses a similar design to previous studies in Norway and England (Bringedal & Feiring, 2011; Miraldo et al., 2014). The questionnaire comprises three sections. Section A collected information concerning attitudes regarding five general statements about personal responsibility and priority setting. Respondents must express their level of agreement with each of the statements on a 5-point Likert-type scale (agree completely, agree partly, neutral, disagree partly, and disagree completely). Section B collected information about five specific responsibility factors that may be considered relevant in priority setting decisions. For each factor, respondents must indicate if (1) it should influence priority setting and (2) the NHS should pay the costs of illness caused by it. The options “Yes,” “No,” and “Don’t Know” were given. Additional data (Section C) included participants’ occupation (health professional or non–health professional), gender, and age. In order to improve the questionnaire’s comprehensibility, it was initially tested on a small sample (10 health professionals and 25 laypersons), and feedback from the pilot testing was incorporated in the final version. For members of the general population, the questionnaire was available for 4 months during 2016 in social networks (Facebook, Linkedin, Twitter, and Google plus) and spread through the personal contacts of the authors. For the health professional sample, the questionnaire was disseminated by e-mail in several public and private health care establishments (hospitals, primary care, clinics, and medical and nurses’ universities).

Following our research protocol, all potential participants were informed about the purpose of this research and the expected duration of participation. We explained that the collection of information was totally anonymous. Participation was voluntary.

This study was approved by a review committee on the protection of human participants.

Participants

The study recorded the opinions of two distinct groups (that fulfill the complete questionnaire)¹: a sample of 295 Portuguese citizens without expertise or experiences in the health care field (throughout named non–health professionals [NHP]) and a sample of 178 health professionals (HP). The group of HP consisted of physicians (55%) and nurses (45%). The majority of respondents (53.9%) were female and 44.6% are older than 44 years.

Data Analysis

Descriptive statistics, the generalized linear models (GLM) and logistic regressions were performed using the SPSS (23). GLM were used to estimate the effects of occupation, age and gender in the level of agreement with each of the five statements about personal responsibility and priority setting (Section A of the questionnaire). GLM is more appropriate to apply than the ordinary linear regression because the dependent variables (personal responsibility statements) do not assume the normal distribution (Nelder & Wedderburn, 1972). Logistic regressions were performed to test for differences in respondents’ opinions concerning the relevance of five risk behaviors in setting priorities between patients and in the funding, by the NHS, of personal-related diseases (Section B of the questionnaire) by occupation, gender, and age. Logistic regressions are more appropriate than the least square models (Wooldridge, 2013).

Results

Section A: General Statements About Personal Responsibility and Priority Setting

Table 1 presents the level of agreement with five statements about personal responsibility (Section A of the questionnaire) of HP and NHP. The results reveal that the HP group seem to agree (partly and completely) more than the NHP group with all the statements. The total level of agreement (partly and completely) of the HP ranged between 42.3% (Statement 2) and 52.4% (Statement 1). The highest level of agreement, of both groups, seems to be with the first statement—“Healthcare priority should depend on the patient’s personal responsibility for the disease.” For HP, the following statement with the greatest agreement was the fifth—“A patient who is responsible for the disease should pay additional co-payments.” The highest level of disagreement (partly and completely) of NHP was with the fourth statement (56.9%)— “Lower priority should be allocated to patients who violate a contract of changes in health-related behaviors” while for HP the level of disagreement with this statement was the lowest. It is worth noting that the number of respondents who were neutral was very low in both groups. Still more NHP than HP declared to be neutral.

Table 1.

Level of Agreement With Personal Responsibility Statements by HP (n = 178) and NHP (n = 295).

Statements about personal responsibility	Disagree completely (%)		Disagree partly (%)		Neutral (%)		Agree partly (%)		Agree completely (%)		Don’t know/No answer (%)
Statements about personal responsibility	NHP	HP	NHP	HP	NHP	HP	NHP	HP	NHP	HP	NHP	HP
Health care priority should depend on the patient’s personal responsibility for the disease	15.2	10.2	24.5	23.7	16.9	13.6	24.1	25.4	16.0	27.1	3.4	0.0
Access to expensive treatment should depend on the patient’s personal responsibility for the disease	19.8	15.3	28.,7	25.4	16.0	11.9	19.8	23.,7	12.2	18.6	3.4	5.1
Access to scarce organ transplants should depend on the patient’s personal responsibility for the disease	20.7	16.9	30.4	27.1	17.7	8.5	19.0	22.0	11.0	25.4	1.3	0.0
Lower priority should be allocated to patients who violate a contract of changes in health-related behaviors	23.6	15.3	33.,3	13.6	18.6	25.4	13.9	25.4	8.4	20.3	2.1	0.0
A patient who is responsible for the disease should pay additional co payments	21.9	18.6	26.6	18.6	24.1	13.6	17.7	25.4	8.0	23.7	1.7	0.0

Note. NHP = non–health professionals; HP = health professionals.

The overall picture is that a majority of the Portuguese HP respondents is, contrary to NHP, absolutely or partly keen to letting personal responsibility for health status influence on the priority setting decisions. On the other hand, the majority of the Portuguese NHP respondents were reluctant in establishing priorities among patients based in health-related behaviors.

Table 2 presents the results of the five linear regressions considering the level of agreement with the personal responsibility statements controlled by occupation, gender, and age. We observe that being a HP (in relation to NHP) increases the level of agreement with letting personal responsibility affect priority setting, except for the first statement (statistically nonsignificant). The extent of the agreement is stronger in the last two statements. For gender, significant differences were found in the second statement, where women respondents agreed more than men that the access to expensive treatments should depend on personal responsibility. Also, some differences were detected according to the respondent’s age. The youngest (≤44 years old) agreed more than the older respondents (>44 years old) that the access to expensive treatments and to organ transplants should depend on personal responsibility for the disease.

Table 2.

Predicting the Level of Agreement With Statements About Personal Responsibility and Priority Setting From Respondents Occupation, Gender, and Age.

Independent variables	Health care priority should depend on the patient’s personal responsibility for the disease	Access to expensive treatment should depend on the patient’s personal responsibility for the disease	Access to scarce organ transplants should depend on the patient’s personal responsibility for the disease	Lower priority should be allocated to patients who violate a contract of changes in health-related behaviors	A patient who is responsible for the disease should pay additional co payments
Constant	1.469	1.322	1.123	1.237	1.358
Occupation	1.601	1.001***	1.077***	2.067**	2.488***
Gender	1.613	0.333**	3.516	1.221	2.084
Age ≤44	0.242	0.047*	2.064**	1.120	2.370
No. of observations	473	473	473	473	473
F test	1.352**	2.818***	3.698***	4.576***	5.858***
R ²	.340	.450	.480	.450	.570

Note. To perform the generalized linear models, the dependent variable assumes values that range between 1 and 5 (Likert-type scale). The independent variables can be summarized as follows: (1) Occupation is a dummy variable that assumes the value 1 if the respondents is a health professionals and 0, if nonhealth professional, (2) Gender is a dummy variable that assumes the value 1 if the respondent is female and 0 otherwise (male), and (3) Age is a dummy variable that assumes the value 1 if the respondents are ≤44 years old and 0 otherwise.

p < .01. ***p < .001.

Section B: Relevance of Five Responsibility Factors in Priority Setting Decisions and in the NHS Funding

Table 3 provide the distribution of responses by HP and NHP about the influence that each of the risky behaviors should have, in deciding whom to treat and, whether NHS should fund treatments caused by each of them (both questions are presented in Section B of the questionnaire). The results suggest that information about illegal drug use, alcohol abuse and smoking was (by this order) considered relevant by both groups of respondents to priority setting decisions. Information about lack of physical exercise was considered less important followed by overweight. Nevertheless, more HP than NHP believed that information concerning overweight/obesity should affect their health care decisions. In general, HP assume that information about all types of risk behaviors more important than NHP except for sedentary lifestyle. These results were reinforced in the second question (bottom part of Table 3). Again, more HP respondents than NHP believe that NHS should not bear the costs of treating illnesses caused by smoking, alcohol, and illegal drug abuse. However, illnesses caused by overweigh/obesity and lack of physical exercise should be funded by the NHS.

Table 3.

Opinions Concerning the Influence of Five Health-Related Behaviors in Setting Priorities and in the Funding of the NHS by Respondents Occupation (HP [n = 178] and NHP [n = 295]).

	Yes		No
Health-related behaviors	NHP	HP	NHP	HP
Should the following health-related behaviors influence the priority for health care?
Smoking	44.7	49.2	49.8	45.8
Overweight/Obesity	19.8	27.1	71.8	64.4
Excessive alcohol	52.7	69.5	41.4	27.1
Illegal drug abuse	58.2	69.5	37.2	23.7
Lack of physical exercise	15.6	8.5	76.8	86.4
Should the NHS pay the costs of the treatments caused by the following health-related behaviors?
Smoking	44.7	39.0	41.8	42.4
Overweight/Obesity	69.6	57.6	17.3	22.1
Excessive alcohol	45.6	32.2	42.2	57.6
Illegal drug abuse	45.1	32.2	41.4	49.2
Lack of physical exercise	67.1	64.4	15.6	11.9

Note. NHS = National Health Service; NHP = non–health professionals; HP = health professionals. The results were presented for “Yes” and “No” and the remainder are “Don’t know.”

To evaluate differences in respondents’ attitudes concerning both questions (Section B of the questionnaire) logistic regressions were applied. By taking the whole sample, we controlled for respondents’ occupation, gender, and age. Tables 4 and 5 present the coefficients and odd ratios (OR) for each regression (question).

Table 4.

Logistic Regressions: Association Between Respondent’s Occupation, Gender, and Age and Their Opinions Concerning the Relevance of Health-Related Behaviors in Prioritizing Patients.

Independent variables	Smoking	Overweight/obesity	Excessive alcohol	Illegal drug abuse	Lack of physical exercise
Occupation	0.393*** (0.371, 1.227)	−0.338 (0.362, 1.404)	0.777* (0.245, 0.861)	0.623* (0.285, 1.010)	0.583 (0.657, 4.886)
Gender	−0.673** (−1.211, 3.172)	−0.080 (0.515, 1.655)	0.227 (0.776, 2.028)	−0.533* (1.041, 2.788)	0.484 (0.829, 3.172)
Age	0.250 (0.731, 2.257)	0.444* (0.304, 1.353)	0.028 (0.586, 1.805)	−0.094 (0.515, 1.611)	−0.145 (0.387, 1.936)
Constant	0.210*	0.923**	0.772*	0.727*	2.488***
No. of observations	473	473	473	473	473
LR χ²(3)	30.23	42.12	45.17	56.24	35.24
Probabilty > χ²	.000	.000	.000	.000	.000
McFadden’s adjusted R²	.400	.3916	.4312	.5612	.4389

Note. LR = likelihood ratio. To perform the logistic regressions, a dummy for the dependent variable was created. This variable assumes the value 1 if respondent answer Yes and 0 otherwise (if answer is No or Don’t know. The independent variables can be summarized as follows: (1) Occupation is a dummy variable that assumes the value 1 if the respondents is a health professionals and 0 otherwise (non–health professional); (2) Gender is a dummy variable that assumes the value 1 if the respondent is female and 0 otherwise (male); and (3) Age is a dummy variable that assumes the value 1 if the respondents are ≤44 years old and 0 otherwise.

p < .05. **p < .01. *** p < .001.

Table 5.

Logistic Regressions: Association Between Respondent’s Occupation, Gender, and Age, and Their Opinions Concerning the Funding by the NHS of Behavior-Related Diseases.

Independent variables	Smoking	Overweight/Obesity	Excessive alcohol	Illegal drug abuse	Lack of physical exercise
Occupation	−0.264** (0.715, 2.370)	−0.429* (0.838, 2.816)	−0.531** (0.916, 3.158)	−0.530** (0.915, 3.156)	0.002 (0.541, 1.857)
Gender	−0.123 (0.549, 1.425)	0.043 (0.626, 1.741)	0.173 (0.738, 1.916)	0.014 (0.628, 1.636)	0.178 (0.720, 1.981)
Age	0.036 (0.592, 1.814)	−0.774* (1.110, 4.235)	−0.069 (0.531, 1.640)	0.126 (0.648, 1.987)	−0.627 (0.986, 3.556)
Constant	−0.426*	0.214**	0.775**	0.763**	0.487*
No. of observations	473	473	473	473	473
LR χ²(3)	20.46	22.46	24.98	29.43	27.34
Probability > χ²	.000	.000	.000	.000	.000
McFadden’s adjusted R²	.3465	.3354	.4221	.4887	.4175

Note. NHS = National Health Service; LR = likelihood ratio. To perform the logistic regressions, a dummy for the dependent variable was created. This variable assumes the value 1 if respondent’s answer is Yes and 0 otherwise (if answer is No or Don’t know). The independent variables can be summarized as follows: (1) Occupation is a dummy variable that assumes the value 1 if the respondents is a health professionals and 0 otherwise (non–health professional); (2) Gender is a dummy variable that assumes the value 1 if the respondent is female and 0 otherwise (male); and (3) Age is a dummy variable that assumes the value 1 if the respondents are ≤44 years old and 0 otherwise.

p < .05. **p < .01. *** p < .001 level.

Table 4 confirms the findings presented in Table 3 by indicating that, HP were more likely, than NHP, to find information about smoking habits (OR = 0.393;confidence interval [CI; 0.371, 1.227]), excessive alcohol consumption (OR = 0.777; CI [0.245, 0,861]) and illegal drug use (OR = 0.623; CI [0.285, 1.010]) relevant to prioritize patients. Respondent’s gender proved to be also significant in explaining the relevance of information concerning two health-related behaviors. Being a woman decreases the probability of finding information about smoking (OR= −0.673; CI [1.211,3.172]) and illegal drug abuse (OR = −0.533; CI [1.041, 2.788]) relevant to priority setting decisions. Regarding age, younger respondents were more likely, than older respondents, to find information about overweight/obesity important in prioritization between patients.

The second regression evaluates if NHS should fund diseases caused by some risk behaviors. Table 5 reveals that being a HP (compared with an NHP) decreases the probability of agreeing with the idea that the NHS should support the costs of treating diseases caused by smoking (OR = 0.264; CI [0.715, 2.370]), alcohol abuse (OR = 0.531; CI [0.916, 3.158]), illegal drug use (OR = 0.530; CI [0.915, 3.156]) and overweight/obesity (OR = 0.429; CI [0.838, 2.816]). Besides occupation, only respondent’s age seems to influence decisions about what the NHS should fund. Younger respondents were more likely, than older respondents, to disagree with the idea that the NHS should pay the costs of treatments caused by overweight/obesity (OR = −0.774; CI [1.110, 4.235]).

These results show that controlling for sociodemographic variables does not alter our initial conclusion, that is, HP were more likely than NHP to support the idea that personal responsibility for disease should count in a priority decision.

Discussion

Personal responsibility (or self-infliction of diseases) has become prominent criteria when discussing priority setting in health care resources, since personal responsibility seems a reasonable criterion to for health care allocation (Buyx, 2008; Schmueli, 2000). This study explores the views of HP about the relevance of health-related behaviors in bedside rationing decisions and compares them with those of laypersons without professional expertise or experience working in the health care field. The results indicate that Portuguese HP found, more than NHP, that information about personal responsibility for disease relevant in priority decisions. A majority of HP expressed a positive attitude in relation to the statement that link priority to personal responsibility. To increase copayments according to personal responsibility for the disease was also particularly popular among HP. These attitudes of Portuguese HP are aligned with the patient’s responsibilities included in the Portuguese Constitution and in the Patient Right Act (Diário da República n, 2014). Among others, the patient has the duty to watch over his state of health. This means that he must seek to participate in the promotion of his own health.

It is worth noting that HP seem very sensitive to the situation in which patients violate the contract they have to change health-related behaviors. We believe that this is a situation that HP often faces in their daily clinical practice. The result for Portuguese HP (safeguarding the differences arising from the sample, the culture and the model of the health system) are quite different from those obtained in Norway (Bringedal & Feiring, 2011). The majority of Norwegian doctors disagreed with the idea that health-related behaviors should be taken into account in a priority decision process (Bringedal & Feiring, 2011). Moreover, a quarter of the Norwegian doctors appear to have no opinion in this matter while for the Portuguese HP this percentage is much smaller.

Among the different types of risk behaviors, both groups of respondents considered information about drug abuse, excessive alcohol consumption and smoking relevant to priority setting decisions (by this order of importance). Considerations regarding lack of exercise and weight were less relevant for the definition of priorities. Overweight/obesity seemed to be the condition considered less relevant when deciding whom to prioritize (although more important for HP), and that may be due, among other factors, to an unhealthy diet and/or lack of exercise. We can only speculate about the reasons of such a lower penalty rate. One possible explanation is that it is commonly known that maintaining a healthy diet is quite expensive preventing numerous people from following it. So, consuming a cheaper diet, rich in fat and sugar, may be an excuse. Future research is needed here. In this regard, it is well known that the adoption of healthy behaviors is influenced by socioeconomic factors (Elo, 2009).

Information about illegal drugs use, excessive alcohol consumption and smoking was considered relevant, by both groups (although more for HP), to priority setting decisions These results are in line with some international findings (Borges & Pinho, 2017; Diederich, Schwettmann, & Winkelhage, 2012; Fortes & Zoboli, 2002; Miraldo et al., 2014). As in our study, Brazilian, English, and German respondents’ disapproval of alcohol abuse was stronger than that of nicotine abuse. However, Norwegian doctors seemed to consider more important considerations about smoking habits than excessive alcohol consumption and the latter more than illegal drug abuse (Bringedal & Feiring, 2011). According to empirical evidence, smoking habits were accepted as a criterion to prioritize patients by a majority of respondents in six countries (Rogge & Kittel, 2016). In relation to obesity/overweighting, Portuguese respondents seem to be aligned with Norwegian doctor’s respondents (Bringedal & Feiring, 2011) and German public (Diederich et al., 2012) but differ from English respondents (Miraldo et al., 2014).

The pattern of Portuguese HP and NHP respondents’ preferences concerning the influence of health-related behaviors in prioritizing between patients seems to be robust and consistent with the preferences revealed concerning the NHS financing of diseases’ treatments resulting from self-harmful health behaviors.

Our findings denote few associations between respondent’s age and/or gender and stated choices. Since women gave less relevance to smoking and illegal drug use than men it seems that they prefer health resources allocation based on need more often than men. Identical findings were obtained elsewhere (Schwappach, 2003). The age of respondents seems to be related only to considerations about obesity/overweight. Information about overweight/obesity was considered more relevant to priority setting decisions by younger respondents than older (older than 44 years). We can only speculate about the reason for this pattern of preferences. One reason may be the growing aesthetics concerns that exist among young people in societies that increasingly value the appearance. There is some research showing that young people perception of overweight and obesity may be very sensitive to public health campaigns, to the weight loss industries and to the media (Friscoe, Houle, & Martin, 2010).

One key strength of our study is the comparison of opinions between Portuguese health professionals and laypersons. A second strength is our use of different types of risk behaviors and the use of different questions that allow testing the robustness of the respondents’ answers. The results should be interpreted with appropriate caution given the nonrandom nature of the samples. The findings cannot be generalized to the Portuguese health care professionals or the Portuguese general public. However, the main goal here was not to achieve an accurate representation of the opinions of the Portuguese people in general but to set up a comparison whereby the value judgments made by two distinct groups could be juxtaposed. Besides this sampling limitation, there are also limitations in using an online questionnaire that raises concerns regarding the quality of the data obtained. Even so, in recent years there has been an increasing interest in collecting data online (Covey, Robinson, Jones-Lee, & Loomes, 2010; Mulhern et al., 2013; Rowen, Brazier, Keetharuth, Tsuchiya, & Mukuria, 2016). The majority of the studies find an overall, broadly similar response throughout all the different survey administration modes (Covey et al., 2010; Mulhern et al., 2013; Rowen et al., 2016). Finally, deciding which patient to treat based on health-related behaviors posed many ethical concerns not the least of it is to know to what extent the risk behaviors contributed effectively to the illness. Furthermore, genetic features and external factors or social determinants to which individuals are exposed to (and are probably beyond their control) such as social class, sheltering conditions, nutrition, working conditions environment, condition in which the individuals’ live, education level, childhood circumstances, and so on, also play a large role in human behavior and consequently are crucial in determining health conditions and lifestyles (Alpinar, Civaner, & Örs, 2010; Bartley, 2004; Marmot & Wilkinson, 2006; Venkatapuram & Marmot, 2009). Differences in childhood circumstances, for example, represent differences in opportunities that are beyond personal responsibility. According to some studies, differences in childhood circumstances such as fetal nutrition, social support, and parental socioeconomic status contribute to inequalities in adult health-related behaviors and health status (Trannoy, Tubeuf, Jusot, & Devaux, 2010; Tubeuf, Jusot, & Bicard, 2012). Childhood circumstances contribute to overall inequality in consumption of fruits and vegetables, nonsmoking and especially obesity (∅vrum & Rickertsen, 2015). Thus, blaming someone, let’s say, a smoker, a heavy drinker, a drug addict and/or an obese for his ailment when he or she is facing a very harsh life circumstances amounts to blame the victims. Blaming the victim is in this case problematic and an ethical issue. Since the present study does not incorporate perceptions about these ethical considerations, our results are mainly of descriptive nature. It is our contention that these drawbacks are overcome by the contribution of this study.

In follow-up research it would be interesting to deepen this study by adding considerations about patients’ social determinants (where they live, level of education, childhood circumstances, where they work, socioeconomic status, etc.). Furthermore, it would be useful to explore attitudes regarding other less explored health-related behaviors, such as excessive sunbathing (the number of skin cancer is raising in Portugal) or dangerous driving (excessive speed, dangerous maneuvers, no respect for signals, etc.). Road accidents are one of the main causes of death (and number of years lost) in the Portuguese population (DGS, 2016).

Overall, our findings suggest that health care professionals seem to have ethical views that are grossly discordant with those of the wider general public they serve. Our study provides suggestive evidence of a possible tension between the opinions of NHS health professionals (those who make rationing decisions in many situations) and laypersons regarding the influence of health responsibility for health status on the priority setting process. This is important insofar as they are the HP who, ultimately, decide whom to treat. Insofar as HP respondents seem to be less reluctant than lay persons to consider information about health-related behaviors in the priority decision process it would be important for the government to carry out health literacy campaigns to encourage people to adopt healthy living habits. Policy makers must focus on a paradigm shift that changes people’s health-related behaviors to avoid disease and achieve healthier years of life and thus reduce inappropriate spending. In this regard, public health preventive measures such as alcohol, smoking, and (more recently) sugar drinking’s taxes are presently implemented in Portugal. These measures, if effective, would increase short-term government income and reduce lifestyle-related mortality.

Differences between the views of public and health professionals raise interesting questions about the role that citizen participation in health decision making should take and how the results of studies such as this should be used. For example, previous evidence suggest that Portuguese population wish to be consulted in matters of rationing health care but do not want the responsibility of having to make those kinds of decisions, that is, they believe doctors should play the most important role in rationing decisions (Botelho, Pinho, & Veiga, 2014). Whoever makes the decisions, it is important to ensure that those decisions are not grossly discordant with the views of members of the public on whose behalf the decisions are being made. Given the increasing pressure on health care budgets and the need to prioritize within health care systems, it becomes important to have knowledge regarding public and professional views on these topics. Only with such knowledge can decision makers develop explicit health care rationing policies and assure that transparency and accountability in medicine are achieved.

Implications for Practice

Allocating health care resources according to personal responsibility for health is highly problematic. This exploratory study suggests that, in Portugal, the views of HP and NHP may differ regarding patient’s prioritization in accordance to health-related behaviors. While HPs consider information about health-related habits relevant to the process of patients’ prioritization, the population cares little about the health consequences of personal risk behaviors. In order to avoid that the opinion of HP prevails and that the popular confidence in them and in the functioning of the NHS are undermined, educational campaigns should be pursued to sensitize lay persons to the harmful effects of certain health related behaviors on owns health status and on public health expenditures. Equally important would be to sensitize HP to the role that social determinants of health can play on health-related behaviors. Policy makers must focus on the following: (1) a paradigm shift that changes people’s health-related behaviors to avoid disease and achieve healthier years of life and thus reduce inappropriate spending and (2) adopting social policies that offer an environment able to promote health in line with the symbiotic relationship that may exist between the individual and social determinants.

Footnotes

Authors’ Note

All procedures were in accordance with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all participants before being included in the study.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

ORCID iD

Micaela Pinho

References

Alpinar

Civaner

Örs

(2010). Rationing healthcare: Should life-style be used as a criterion? Anatolian Journal of Cardiology, 10, 367-371.

Bartley

(2004). Health inequality: An introduction to theories, concepts and methods. Cambridge, England: Polity Press.

Borges

Pinho

(2017). Should lifestyles be a criterion for healthcare rationing? Evidence from a Portuguese survey. Journal of Research in Health Sciences, 17(4), 1-7.

Botelho

Pinho

Veiga

(2014). Who should participate in health care priority setting and how should priorities be set? Evidence from a Portuguese survey. Revista Portuguesa Saúde Pública, 31, 214-222.

Bringedal

Feiring

(2011). On the relevance of personal responsibility in priority setting: A cross-sectional survey among Norwegian medical doctors. Journal of Medical Ethics, 37, 357-361.

Buyx

(2008). Personal responsibility for health as a rationing criterion: Why we don’t like it and why maybe we should. Journal of Medical Ethics, 34, 871-874.

Callahan

Koenig

Minkler

(1999). Promoting health and preventing diseases: Ethical demands and social challenges. International Quarterly of Community Health Education, 18, 163-180.

Cappelen

Norheim

(2005). Responsibility in health care: a liberal egalitarian approach. Journal of Medical Ethics, 31, 476-480.

Covey

Robinson

Jones-Lee

Loomes

(2010). Responsibility, scale and valuation of rail safety. Journal Risk Uncertain, 40, 85-108.

10.

Daniels

Sabin

(1997). Limits to health care: Fair procedures, democratic deliberation, and the legitimacy problem for insurers. Philosophy and Public Affairs, 26, 303-350.

11.

DGS. (2016, December). Revista Científica da Direção-Geral da Saúde [Health by numbers]. Proceedings of the 1st Portuguese National Public Health Congress. Retrieved from https://www.dgs.pt/paginas-de-sistema/saude-de-a-a-z/revista-da-dgs.aspx

12.

Diário da República n. (2014). Lei consolidando a legislação em matéria Direitos e deveres do utente dos serviços de saúde. [Law consolidating the legislation on the rights and duties of the user of health services]. Retrieved from https://dre.pt/web/guest/pesquisa/-/search/571943/details/maximized

13.

Diederich

Schwettmann

Winkelhage

(2012). Does lifestyle matter when to decide on co-payment for health care? Attitudes of the German public (Priorisierung in der Medizin FOR 655 Nr. 35/012 working paper series). Retrieved from https://pdfs.semanticscholar.org/2b50/814b5ceb054d7f2a6e2903ba1cf6fb354148.pdf

14.

Elo

(2009). Social class differentials in health and mortality: Pattern and explanations in comparative perspectives. Annual Review Sociology, 35, 553-572.

15.

Eurostat. (2017). Eurostat, Statistics explained—Health in the European Union: Facts and figures. Causes of death statistics 2017. Retrieved from http://ec.europa.eu/eurostat/statistics-explained/index.php/Health_in_the_European_Union_%E2%80%93_facts_and_figures

16.

Exel

Baker

Mason

Donaldson

Brouwe

(2015). Public views on principles for health care priority setting: Findings of a European cross-country study using Q methodology. Social Science & Medicine, 126, 128-137.

17.

Fortes

Zoboli

(2002). A study on the ethics of microallocation of scarce resources in health care. Journal of Medical Ethics, 28, 266-269.

18.

Friscoe

Houle

Martin

(2010). The image in the mirror and the number on the scale: Weight, weight preferences, and adolescent depressive symptoms. Journal of Health and Social Behavior, 51, 215-228.

19.

Lancsar

Ghijben

Butler

Donaldson

(2015). Attributes and weights in healthcare priority setting: A systematic review of what counts and to what extent. Social Science & Medicine, 146, 41-52.

20.

Jepson

Harris

Platt

Tannahill

(2010). The effectiveness of interventions to change six health behaviors: A review of reviews. BMC Public Health, 10, 538. doi:10.1186/1471-2458-10-538

21.

Lopes

Torres

Oliveira

Severo

Alarcão

Guiomar

. . . Consórcio, IAN-AF. (2017). Inquérito Alimentar Nacional e de Atividade Física 2015-2016: Relatório metodológico [National food and physical activity survey 2015-2016: Methodological report]. Retrieved from https://ian-af.up.pt/sites/default/files/IAN-AF%20Relatorio%20Metodol%C3%B3gico.pdf

22.

Marmot

Wilkinson

(2006). Social determinants of health. Oxford, England: Oxford University Press.

23.

Minkler

(1999). Personal responsibility for health? A review of the arguments and the evidence of century’s end. Health Education Behavior, 26, 121-140.

24.

Miraldo

Galizzi

Merla

Levaqqi

Schulz

Auxilia

. . . Gellati

(2014). Should I pay for your risky behaviours? Evidence from London. Preventive Medicine, 66, 145-158.

25.

Mulhern

Longworth

Brazier

Rowen

Bansback

Devlin

Tsuchiya

(2013). Binary choice health state valuation and mode of administration: Head-to-head comparison of online and CAPI. Value in Health, 16, 104-113.

26.

Nelder

Wedderburn

(1972). Generalized linear models. Journal of the Royal Statistical Society: Series A (General), 135, 370-384.

27.

OECD. (2015). OECD Health Statistics 2015. Retrieved from https://www.oecd.org/unitedstates/Country-Note-UNITED%20STATES-OECD-Health-Statistics-2015.pdf

28.

Olsen

Richardson

Dolan

Menzel

(2003). The moral relevance of personal characteristics in setting health care priorities. Social Science & Medicine, 73, 1163-1172.

29.

∅vrum

Rickertsen

(2015). Inequality in health versus inequality in lifestyles. Nordic Journal of Health, 1, 18-33.

30.

Pinho

Borges

Zahariev

(2017). Bedsides healthcare rationing dilemmas: A survey from Bulgaria and comparison with Portugal. Social Theory & Health, 15, 285-301.

31.

Pordata. (2018). Portugal, População [Portugal population]. Fundação Francisco Manuel dos Santos. Retrieved from https://www.pordata.pt/Portugal

32.

Rogge

Kittel

(2016). Who shall not be treated: Public attitudes on setting health care priorities by person-based criteria in 28 nations. PLoS One, 11, e0157018.

33.

Rowen

Brazier

Keetharuth

Tsuchiya

Mukuria

(2016). Comparison of modes of administration and alternative formats for eliciting societal preferences for burden of illness. Applied Health Econ Health Policy, 14(1), 89-104.

34.

Sassi

Hurst

(2008). The prevention of lifestyle-related chronic diseases: An economic framework (OECD Health Working Paper No. 32). Retrieved from http://www.oecd.org/health/health-systems/40324263.pdf

35.

Schmidt

(2009). Just health responsibility. Journal Medical Ethics, 35, 21-26.

36.

Schmidt

Asch

Halpern

(2012). Fairness and wellness incentives: What is the relevance of the process-outcome distinction? Prevention Medicine, 55(Suppl.), 5118-5123.

37.

Schmidt

Voigt

Wikler

(2010). Carrots, sticks, and health care reform: Problems with wellness incentives. New England Journal of Medicine, 362, e3 doi:10.1056/NEJMp0911552

38.

Schmueli

(2000). Horizontal equity in medical care: A case study of Israeli public’s views. Israel Medical Association Journal, 2, 746-752.

39.

Schwappach

(2003). Does it matter who you are or what you gain? An experimental study of preferences for resource allocation. Health Economics, 12, 255-267.

40.

SICAD. (2016). Relatório Anual 2015: A situação do país em matéria de álcool [Annual report 2015: The state of the country in alcohol matters]. Retrieved from http://www.sicad.pt/PT/Publicacoes/Documents/Relat%C3%B3rio%20Anual%20%C3%81lcool%20%202015.pdf

41.

Stegeman

Willems

Dekker

Bossuyt

(2014). Individual responsibility, solidarity and differentiation in healthcare. Public Health Ethics, 40, 770-773.

42.

Trannoy

Tubeuf

Jusot

Devaux

(2010). Inequalities of opportunities in health in France: A first pass. Health Economics, 19, 921-938.

43.

Trappenburg

(2000). Lifestyle solidarity in the healthcare system. Health Care Analysis, 8, 65-75.

44.

Tubeuf

Jusot

Bicard

(2012). Mediating role of education and lifestyles in the relationship between early-life conditions and health: Evidence from the British cohort. Health Economics, 21(Suppl.), 129-150.

45.

Venkatapuram

Marmot

(2009). Epidemiology and social justice in light of social determinants of health research. Bioethics, 23(2), 79-89.

46.

World Health Organization. (2017). Noncommunicable diseases. Retrieved from http://www.euro.who.int/en/health-topics/noncommunicable-diseases

47.

Wooldridge

(2013). Introductory econometrics: A modern approach (5th ed.). Mason, OH: South-Western Cengage Learning.