Just responsibilities? On responsibility for health in Swedish healthcare priority setting

Abstract

The overarching aim of this project is to explore, empirically as well as philosophically, the arguments for and against making Swedish healthcare priority setting decisions sensitive to patients’ degree of responsibility for their ill health. Arguments of interest are those expressed by important stakeholders in the debate – physicians, lay people and, as this is a matter of some theoretical importance, bioethicists.

Article I: An experimental vignette methodology study. The two versions of the vignette differed in one aspect: the patient was a non-smoker/patient was a smoker. The aim was to investigate whether the willingness to offer a novel, expensive and moderately life prolonging treatment to one patient depends upon whether this patient was a non-smoker or a smoker. Sample: 1193 physicians and 962 citizens. Results: Response rate 50,9%. Respondents were more willing to offer a novel, expensive and moderately life prolonging treatment if the patient was a non-smoker than a smoker. This statistically significant difference in willingness to offer treatment was irrespective of considerations of capacity to benefit, and held among citizens (83,8% vs 68,7%), oncologists (81,8% vs 64,9%) and GPs (69,1% vs 56,3%), but not among pulmonologists (72,9% vs 67,2%). Conclusion: Members of the citizenry and most physician specialities are more willing to offer treatment to a non-smoker than to a smoker. This is in conflict with the Swedish Ethical Platform for priority setting. Somewhat paradoxically, the lower willingness to offer treatment to the smoking patient is unrelated to considerations of responsibility among physician respondents (but not among citizens).

Article II: An experimental vignette methodology study using parts of the data material from Article I (only physician respondents). The aim was to investigate whether there is a difference in the perception of a medical indication for treatment of a non-smoking or smoking patient depending upon the respondent’s own attitude towards the treatment (as measured by whether the respondent’s trust in healthcare would change – for better or worse – if treatments such as this were to become routine). Sample: 1193 physicians. Results: Response rate 54,1%. Only physicians whose trust would change differed in their willingness to offer the treatment depending upon whether the patient was a non-smoker or a smoker (78% vs 57%). Among physicians whose trust would be unchanged there was no corresponding statistically significant difference (70% vs 67%). Similarly, only physicians whose trust would change based their judgment of whether it was medically indicated to provide the treatment (in part) on the patient’s smoking status. These physicians judged treatment “medically indicated” more often in the case of the non-smoking patient than the smoking patient (67% vs 50%). Among physicians whose trust would be unchanged there was no corresponding difference (53% vs 53%). Among all categories of physicians the notion of “medical indications” was used in a widely differing ways. Conclusion: Physicians whose trust in healthcare would be affected by a change in treatment routine are more likely to base treatment decisions, as well as judgments of medical indication, on medically irrelevant factors. Physicians seem to use the term “medical indications” in an ambiguous manner.

Article III: A vignette-based questionnaire study. The aim was to investigate the support for making planned surgery conditional upon smoking cessation prior to surgery, as well as the reasons given in support or rejection of this. Sample: 795 physicians and 485 citizens. Results: Response rate 56,6%. Most respondents (physicians as well as citizens) agreed that planned surgery should be made conditional upon the vignette patient’s smoking cessation (83,9% and 86,6%, respectively). Additional findings include that the most common reason for supporting a demand for smoking cessation was the (assumed) increased peri-operative risks, but that another weighty consideration was that smoking cessation would bring about long-term positive effects. Finally, analysis of the written comments indicated that the support referenced above may to a great extent be for the less harsh “recommendation of smoking cessation” policy rather than for the harsher “requirement of smoking cessation” policy. Conclusion: There is a strong support for some kind of policy of smoking cessation prior to planned surgery, but it is unclear whether respondents support a policy of requiring or merely recommending smoking cessation. Reasons for supporting this policy include the peri-operative risks as well as long-term health gains. There is risk that paternalistic sentiments influence support for policies of smoking cessation.

Article IV: A philosophical (non-empirical) article. The aim was to discuss several possible challenges when applying Luck Egalitarianism as a guide for healthcare policy, using considerations that appear reasonable and relevant even to Luck Egalitarians themselves. Conclusion: Luck Egalitarianism faces great challenges when applied to the healthcare setting. Some of these challenges may be overcome: for instance, it may be possible to define “imprudent actions” so that Luck Egalitarianism escapes charges of moral arbitrariness. Other challenges including how to decide between the ex ante and ex post operationalisations of Luck Egalitarianism, and setting the threshold for abandonment, are more difficult to overcome. Any way of responding to these challenges are likely to be unattractive even to Luck Egalitarians themselves.

Article V: An interview study using the phenomenographic method of data analysis. The aim was to explore physicians’ understanding of the notion of a personal responsibility for health. Sample: 14 General Practitioners. Results: All participants found the notion of personal responsibility for health relevant to their clinical practice. There was a wide range of understandings of this notion. The understandings differed as to the perceived origins of responsibility as well as the expressions of responsibility. The GP herself was perceived as a key player in shaping and defining the construct of patients’ personal responsibility for health. Considerations of patients’ personal responsibilities may trigger strong emotional reactions in GPs. Conclusion: The notion of personal responsibility for health is relevant but complex, and discussions about personal responsibility for health are emotionally challenging.