Exploring the Influence of Decision-making Style and Financial Status of Cancer Patients on Varied Cultural Aspects of Decision Process

Abstract

Though many studies have linked shared decision-making with positive patient outcomes and patient satisfaction, there is less research on the practicality and feasibility of such an approach, specifically, in India. Recent findings indicate that contextual constraints of the patients and family caregivers are fused with and inevitably shape their decision-making style. This study investigates the influence of individual (paternalistic and shared decision-making style) and contextual (financial status of the patients) factors on certain cultural aspects of decision process, namely, trust in doctors, patient’s agency, emotional distress and attribution of responsibility. Data were collected using a survey questionnaire from a sample of 306 participants. Linear mixed models were used to analyse the responses. The results indicate that a shift towards shared approaches does yield qualitatively superior outcomes in terms of increased trust in doctors and patients’ agency along with reduced emotional distress. The study also emphasises the role of cultural metaphysical beliefs in attributing responsibility for treatment decisions. Further, low financial status was found to be associated with higher trust in doctors, lower patient agency, and increased emotional distress for both patients and family caregivers. The move towards shared decision-making, considering the contextual realities of patients, could be instrumental in addressing critical issues, such as prevalence of collusion, superficial role of informed consent, and high reliance on doctors’ authority or expertise.

Keywords

Culture decision process context shared decision-making cancer

Introduction

With advances in technologies, both doctors and patients have huge amounts of information at hand. This information makes the decision process more complex (Lantos, 2021). In the literature on doctor–patient relationship and interaction, there is a move towards evidence-based, informed, and shared decisions (Gillman & Ferrer, 2021; Sankar et al., 2018). Though there are many studies that have linked such decisions with positive patient outcomes and patient satisfaction (Entwistle et al., 2012; Gillman & Ferrer, 2021; Sankar et al., 2018), there is less research on the practicality and feasibility of such an approach (Ozdemir & Finkelstein, 2018), specifically, in India (Gopichandran, 2019; Sankar et al., 2018). There are certain unique aspects that define doctor–patient relationship in India, such as reverence towards authority (Panda, 2023; Subramani, 2019), paternalistic family dynamics (where often the decision-maker is the head of the family) (Gopichandran, 2019; Subramani, 2019), lower agency of female patients than their male counterparts (Bhakuni, 2020; Gopichandran, 2019), power distance between doctors and patients (Jacob, 2014), substantially low levels of cancer awareness among public (Chandra et al., 1998), among others. Research exploring how doctors and patients communicate to reach shared and informed decisions during the treatment process given such conditions is conspicuously deficient (Bhakuni, 2020; Gopichandran, 2019; Panda, 2023; Sankar et al., 2018). The doctor–patient communication is also not immune to the sociocultural influences, such as beliefs of the patients and the family caregivers regarding full disclosure of information, attribution of decision responsibility to metaphysical factors, and the extent of trust in the medical fraternity (Chaturvedi et al., 2009; Chittem et al., 2021; Mathew et al., 2021; Victor et al., 2018).

Studies suggest that in India, the patients and their families often put the burden of the medical decision solely on the doctors’ expertise and remain a passive audience to the decision-making process (Jacob, 2014; Sareen & Dutt, 2019). Various factors such as high financial burden of cancer, low education and knowledge about the disease, and low financial status that renders the possibility of taking second opinions impossible, were associated with the high prevalence of paternalistic decision-making (Harding et al., 2022; Mathew et al., 2020). Conversely, patients with higher socioeconomic status and greater literacy levels were more actively involved in the decision process (Datta et al., 2017; Harding et al., 2022). Such findings indicate that contextual constraints of the patients and family caregivers are fused with and inevitably shape their decision-making style (Blok et al., 2023; Datta et al., 2017; Ram Prakash & Lingam, 2021; Zafar & Peppercorn, 2023).

The present study focuses on four crucial aspects of the decision process: trust on doctors, patients’ agency, emotional distress and attribution of responsibility. The study examines how individual (paternalistic and shared decision-making style) and contextual (financial status of the patients) factors influence certain aspects of decision process, namely, trust in doctors, patient’s agency, emotional distress and attribution of responsibility? The aim was not to assess the participants’ actual decisions or evaluate the accuracy/optimality/efficiency of the decision outcome or analyse the strategies/heuristics used to make the decision (Klein, 2017; Weiss & Shanteau, 2021), rather, through hypothetical vignettes simulate the experience of making decisions during cancer treatment, considering factors, such as socioeconomic status and decision-making style.

The Contradiction of Shared Decision-making

In the last decade, shared decision-making has emerged as the preferred approach for healthcare professionals, patients and policy makers, while paternalistic approach is increasingly perceived as outdated and potentially detrimental for patient care (Driever et al., 2021; Lantos, 2021; Thompson & Whiffen, 2018). In the shared decision-making approach, medical decisions are considered a collaborative outcome involving both patients and doctors and are therefore believed to be more aligned with the contextual realities and values of patients (Lantos, 2021; Pollard et al., 2015). However, recent studies have challenged the feasibility and effectiveness of shared decision-making (Driever et al., 2020, 2021; Hahlweg et al., 2017; Oprea et al., 2023).

In a cross-sectional survey of 785 doctors, Driever et al. (2020) found that though most doctors preferred shared decision-making, in their everyday practice they often leaned towards paternalistic decision-making (Oprea et al., 2023). Physicians level of support for shared decision-making was found to be associated with various factors such as uncertainty in evidence (Driever et al., 2021; Pollard et al., 2015), time pressure (Hahlweg et al., 2017), severity of treatment decisions (Pollard et al., 2015), patient characteristics, such that more active patients facilitated shared decision-making (Hahlweg et al., 2017; Pollard et al., 2015). In a subsequent study by Driever et al. (2021), it was found that medical residents often confused shared decision-making with obtaining informed consent and/or letting patients decide when faced with inconclusive evidence.

Research examining patients’ preferences for medical information and their desired role in decision-making has yielded inconsistent findings (Elkin et al., 2007; Josfeld et al., 2021; Sankar et al., 2018; Thompson & Whiffen, 2018). Josfeld et al. (2021) surveyed 220 cancer patients and found that participants desired more information than they received, often feeling overwhelmed with information or underinformed, particularly during consultations with doctors. The majority expressed a preference for shared decision-making over deciding alone, with none indicating a preference for doctors making decisions on their behalf (Josfeld et al., 2021). In contrast, Sankar et al. (2018) found that while most participants expressed a strong desire to know their diagnosis and prognosis, 97% (of the 133 cancer patients surveyed in India) wanted their doctors to make decisions regarding treatment. Another study involving 73 elderly cancer patients found that only 44% wanted survival-related information, with approximately half preferring a passive role in decision-making (Elkin et al., 2007). Recently, some researchers have challenged the Western emphasis on patient autonomy, suggesting that paternalism may hold value in other cultural contexts for delivering high-quality ethical care (Heidenreich et al., 2017; Papathanassoglou et al., 2012; Thompson & Whiffen, 2018).

With such mixed evidence regarding prevalence and feasibility of the shared decision-making approach, this study takes a step back to investigate whether the two decision-making styles indeed differ in their association with core aspects of the decision process. Further, the study incorporates the contextual dimension of patients’ financial status to further enrich the understanding of decision-making styles.

Core Aspects of Decision Process

The present study identified four core aspects of the decision process: trust in doctors, patient’s agency, emotional distress and attribution of responsibility. Previous literature suggests that as shared decision-making acknowledges patients’ autonomy (Guadalajara et al., 2022; Lantos, 2021), it fosters greater trust in doctors for both patients and caregivers (Entwistle, 2004; Lincoln et al., 2023) and elicits higher patient satisfaction (Oprea et al., 2023; Pollard et al., 2015). Financial toxicity is also found to be linked to higher emotional distress (Abrams et al., 2021; Carrera et al., 2018), lower agency in decision process (Boby et al., 2021; Prinja et al., 2023) and higher trust in doctors (Abrams et al., 2021; Boby et al., 2021; Zafar & Peppercorn, 2023). Studies examining cultural beliefs in the face of illness also suggest that patients often attribute the cause of their illness to metaphysical factors, such as god, fate or karma (Agrawal & Dalal, 1993; Dalal & Misra, 2006; Kohli & Dalal, 2016).

The following hypothesis was tested:

H₁: Trust on doctors would be higher for conditions of the shared decision-making as compared to those of paternalistic decision-making.

H₂: Patient’s agency would be higher for conditions of the shared decision-making as compared to those of paternalistic decision-making.

H₃: Emotional distress for patients and family caregivers would be lower for conditions of the shared decision-making as compared to those of paternalistic decision-making.

H₄: Attribution of responsibility to doctors, god and chance/luck would be lower and to patients and family caregivers would be higher for conditions of the shared decision-making as compared to those of paternalistic decision-making.

H₅: Trust on doctors would be higher for conditions of low financial status as compared to high financial status.

H₆: Patient’s agency would be lower for conditions of low financial status as compared to high financial status.

H₇: Emotional distress for patients and family caregivers would be higher for conditions of low financial status as compared to high financial status.

H₈: Attribution of responsibility to doctors, god and chance/luck would be higher and to patients and family caregivers would be lower for conditions of low financial status as compared to high financial status.

H₉: There would be a significant interaction effect between decision-making style (shared versus paternalistic) and financial status (low versus high) on trust in doctors, patient’s agency, emotional distress and attribution of responsibility (doctors, patients, family caregivers, god and chance/luck).

Method

Study Design

A 2 (decision-making style: paternalistic, shared) × 2 (financial status: high, low) within-subject factorial design was used to examine the effect of the two independent variables on four dependent variables: trust on doctors, patient’s agency, emotional distress in patient and family caregivers, and attribution of responsibility. Hypothetical vignettes were constructed on the basis of the four conditions:

High financial status and paternalistic decision-making

Low financial status and paternalistic decision-making

High financial status and shared decision-making

Low financial status and shared decision-making

The participants responded to all four vignettes and, therefore, a within-subject random error was introduced. Linear mixed models (LMMs) were used for analysis as the data had both fixed effects (decision-making style and financial status) and random effects (variation due to subjects). The design was fully crossed and balanced, subject × condition, such that all subjects were part of all four conditions. LMMs are often considered a more robust technique (as compared to repeated measures ANOVA) in accounting for such error variation (Walker et al., 2019).

Sample

The sample consisted of 306 participants residing in various parts of India, mainly from, Delhi, NCR (National Capital Region), Uttarakhand and Haryana. The mean age of participants was 34.02 years with a range of 17–72 years. The demographic details are depicted in Table 1. Data were collected using a Google form that was circulated to the participants using a combination of convenience and snowball sampling.

Table 1.

Sociodemographic Characteristics of Participants.

Demographic Variables	Participants (n = 306)
Age
Below 30 years	159
31–64 years	133
65 years above	4
Gender
Males	133
Females	173
Socioeconomic class
Lower	57
Upper	249
Education level
Professional	21
Postgraduate	134
Undergraduate	115
High school	26
10th standard	6
5th–8th standard	3
None	1

Inclusion/Exclusion Criteria

Participants who had been diagnosed with cancer or had acted as a primary caregiver for a cancer patient were purposely excluded from the sample. It was because their previous experience of taking decisions under such conditions would have acted as a major confounding variable. Forty-eight such participants were identified and were excluded from the analysis, leaving a sample of 258 participants.

Measures

A survey questionnaire was constructed that consisted of a brief detail about the study with informed consent, followed by a section on demographic information and four hypothetical vignettes, each followed by the measures for dependent variables.

Demographic Information

The survey included a section seeking demographic information from participants regarding, age, gender, level of education, their perceived socioeconomic status, area of residence, and inclusion/exclusion criteria.

Hypothetical Vignettes

Four hypothetical vignettes were constructed based on the unique combinations of the independent variable. The vignettes were drafted meticulously to isolate the effects of the independent variables while maintaining rigorous control over extraneous factors. Gender-neutral language was consistently employed, and potential confounding factors such as age, place of residence and religious inclinations were deliberately excluded from the vignettes. Throughout, language was intentionally kept clear and free of technical jargon. The vignettes were constructed to reflect real-world variations.

To manipulate financial status, two scenarios were developed. In the high financial status conditions, the character ‘X’ was portrayed as employed in a big corporate company with more than sufficient savings. Conversely, in the low financial status conditions, ‘X’ assumed the role of a manual labourer, with their family’s savings being meagre. Importantly, all vignettes explicitly conveyed that ‘X’ lacked medical insurance, eliminating potential confounding effects.

To manipulate decision-making style, the different aspects of paternalistic and shared decision-making were included to capture their meaning. In the paternalistic decision-making conditions, ‘X’ and their family members pushed the physician to make treatment decisions, emphasising their trust in the doctor’s expertise. In contrast, the shared decision-making conditions depicted ‘X’, their family members and the doctor engaging in collaborative discussions to jointly determine treatment options.

Dependent Variable Measures

The measures for dependent variable were based on the findings of Study 1 and 2 of the thesis as well as previous literature.

Trust in doctors: This was assessed by developing a comprehensive measure consisting of six distinct dimensions. The first dimension, competence, evaluated the extent to which participants trusted the expertise and competence of the doctors (Hall, 2006). The second dimension gauged the level of respect for doctors’ authority (Gopichandran, 2019). The third dimension, fidelity, delved into the honesty of communication with patients and the respect for their agency (Gopichandran, 2019; Hall, 2006). Loyalty, the fourth dimension, focused on the loyalty exhibited towards doctors who had provided past treatment (Gopichandran, 2019). The fifth dimension, experience level, considered the seniority and experience of the doctors. Finally, the sixth dimension, global trust, captured a holistic sense of trust encompassing all dimensions (Hall, 2006). The measure thus included six items measured on a 5-point Likert scale, ranging from 1 (no trust) to 5 (complete trust). The total score for trust in doctors was computed by summing the scores for each of the six items.

Patients’ agency: This included a single item measured on a 5-point Likert scale, ranging from 1 (no agency) to 5 (complete agency): In the given scenario, how much do you think the patient has a voice or say in deciding the best treatment option?

Emotional distress: The perceived experience of emotional distress for both patient and family caregivers in the vignettes was operationalised in terms of five emotions: mental distress, helplessness, fear, anxiety and anger. The five items were measured on a 5-point Likert scale and the total score for emotional distress for both patient and family caregivers was calculated by summing the individual scores of all five items.

Attribution of responsibility: This variable was operationalised in terms of blame or accountability that would be placed on doctors, patients, family caregivers, god and chance or luck, in case the chosen treatment option in the vignette severely hampered the patients’ quality of life and was unable to eliminate cancer. The participants rated the amount of responsibility to each of the five factors on a 5-point Likert scale ranging from 1 (no responsibility) to 5 (complete responsibility).

Procedure

The survey questionnaire was translated into Hindi language and then back translated to English by an expert. A pilot study was conducted to ensure the clarity and coherence of the questionnaire on a sample of 14 participants. The age range of the participants was 26–60 years. In the pilot study, the participants demonstrated a satisfactory understanding of the questionnaire items; however, some participants expressed a need for enhanced clarity regarding the hypothetical vignettes. On the basis of this feedback, a color-coding system was implemented to accentuate the distinctions between the four vignettes. This visual enhancement aimed to facilitate participants’ comprehension and engagement with the scenarios, ensuring that the nuances of the experimental conditions were readily discernible. These Google forms were then circulated online to collect data.

Analysis

The statistical analysis was conducted using R program, version 4.2.3 (RStudio Team, 2022). LMMs were used to analyse responses to the four vignettes. There were two fixed effects (decision-making style and financial status) and one random effect (within-subject factor). The analysis aimed to calculate the main effects and interaction effect of the fixed effects on the four dependent variables while controlling for the random effect. Conducting LMM analysis included:

Understanding the model structure: The model was created as follows:

Model 1: Trust in doctors ~ Decision-making × Financial status + (1 + Decision-making + Financial status|Subject)

Dependent variable: Trust in doctors

Fixed effects:

Decision-making style (DM) at two levels: paternalistic (P) and shared (S)

Financial status (FS) at two levels: high (H) and low (L)

Four conditions:

HP: High FS and paternalistic DM

LP: Low FS and paternalistic DM

HS: High FS and shared DM

LS: Low FS and shared DM

Random effect: Within-subject variation understood in terms of random intercept for subjects and random slopes for decision-making and financial status at each level of subject. Random intercept (1|Subject) assumes that the effects of decision-making style and financial status are the same across all subjects. In other words, the difference in trust in doctors due to decision-making style and financial status would be the same for all the subjects. By adding random slopes to the model (Decision-making + Financial status|Subject), variations among the subjects due to the fixed effects are allowed. For instance, the influence of the two fixed effects might lead to a more significant 2-unit increase in trust in doctors for some participants, while for others, it may lead to only a 0.5-unit increase. To account for such variation, random slopes were added to the model (Singmann & Kellen, 2019; Sparks, 2018).

Setting contrasts: Sum-to-zero contrasts were set wherein the intercept corresponds to the grand mean as compared to treatment contrasts where the intercept corresponds to the first group (Howell, 2019).

Sum-to-zero Contrast for Linear Mixed Modelling.

	1	2	3
P_H	1	0	0
P_L	0	1	0
S_H	0	0	1
S_L	−1	−1	−1

Model selection: To check whether adding random slopes helped to better explain the model or not, another model with only the random intercept for subjects was created:

Model 2: Trust in doctors ~ Decision-making × Financial status + (1|Subject)

To check goodness-of-fit, Akaike information criterion (AIC) was used to select the model. The lower the AIC, the better the model fit (Singmann & Kellen, 2019; Walker et al., 2019). Therefore, Model 1 with slightly lower AIC = 2,451 as compared to Model 2 with AIC = 2,474, was chosen.

Checking model assumptions: Model 1 was then checked for assumptions using performance package (Lüdecke et al., 2021). The assumptions of normality of residuals, homogeneity of variance, linearity and collinearity were examined. The package also provides additional measures to ensure the robustness of model, such as posterior predictive check (whether model predictions resemble observed data) and presence of influential observations (outliers with leverage to influence the regression slope). In case the assumptions were not satisfactorily met, data transformation was conducted according to the recommendations of Howell (2019). Data transformation involves reshaping the data in a manner that enhances its compatibility with regression analysis, thereby improving the model’s performance and accuracy.

Summary of model and interpretation: After checking the model and making necessary changes, the LMM analysis was conducted using lme4 package (Bates et al., 2023) and car package (Fox et al., 2023). The output and interpretation would be discussed in the results section.

Checking effect size: Beyond the statistical significance, it is important to estimate the practical significance or the meaningfulness of the observed results. For LMMs, intra-class correlation coefficients (ICC) or pseudo-R-squared are considered good estimates for effect size (Nakagawa et al., 2017). Both ICC and pseudo-R-squared give roughly the same estimate and help in understanding the amount of variance explained by the model. The variation in LMMs comes from multiple sources (fixed effect, random effect and intercept). Though these estimates measure the amount of variation attributed to fixed effects, it is not possible to further dissect the variance and account for variation due to each particular fixed effect (Nakagawa et al., 2017). Thus, pseudo-R-squared was calculated using MuMIn package to understand the amount of variance explained by the entire model and by the fixed effects.

Post-hoc comparisons: For post-hoc comparisons, EMMEANS package (Lenth, 2023) was used that employs the Tukey method for pairwise comparisons.

Graphical presentation of data: The ggplot2 package (Wickham et al., 2023) was used to graphically present the main and interaction effects of the fixed effects on the response variables.

These steps were followed for all response variables. In total, nine LMMs [trust in doctors (1), patient’s agency (1), emotional distress for patients (1), emotional distress for family caregivers (1) and five factors of attribution of responsibility (5)] were created and interpreted.

Results

Tables 3 and 4 summarise the results of the LMMs for all four dependent variables.

Table 2.

Descriptive Statistics for the Four Dependent Variable Across All Four Conditions.

	Dependent Variables	HP		LP		HS		LS		Total
	Dependent Variables	M	SD	M	SD	M	SD	M	SD	M	SD
1.	Trust in doctors	22.57	5.73	24.09	5.25	23.84	4.96	24.19	5.16	23.67	5.31
2.	Patient’s agency	3.65	1.21	3.15	1.37	3.71	1.22	3.26	1.36	3.44	1.31
3.	Emotional distress
(i)	Patients	17.64	4.64	19.10	5.09	16.50	5.35	18.18	5.27	17.86	5.17
(ii)	Family caregivers	17.53	4.43	19.07	4.98	16.38	5.26	18.14	5.18	17.78	5.06
4.	Attribution of responsibility
(i)	Doctors	3.01	1.35	3.15	1.46	2.95	1.42	3.06	1.42	3.04	1.41
(ii)	Patients	2.25	1.28	2.19	1.28	2.35	1.32	2.21	1.27	2.25	1.29
(iii)	Family caregivers	2.41	1.30	2.31	1.31	2.49	1.37	2.41	1.31	2.40	1.33
(iv)	God	2.68	1.33	2.81	1.53	2.65	1.46	2.80	1.52	2.74	1.52
(v)	Chance/luck	3.24	1.47	3.27	1.45	3.14	1.47	3.17	1.48	3.21	1.46

Table 3.

Summary of Linear Mixed Model for Trust in Doctors, Patient’s Agency and Emotional Distress.

	Dependent Variable
	Trust in Doctor	Patient’s Agency	Emotional Distress_P	Emotional Distress_F
	(1)	(2)	(3)	(4)
Decision-making	−0.050**	−0.018*	0.015***	0.016***
	(0.016)	(0.008)	(0.003)	(0.003)
	t = −3.149	t = −2.324	t = 4.750	t = 5.367
	p = .002	p = .021	p = .00001	p = .00000
Financial status	−0.089***	0.071***	−0.020***	−0.021***
	(0.019)	(0.009)	(0.003)	(0.003)
	t = −4.664	t = 7.478	t = −6.939	t = −6.840
	p = .00001	p = .000	p = .000	p = .000
Decision-making: Financial status	−0.037**	0.002	0.004	0.004
	(0.013)	(0.006)	(0.002)	(0.002)
	t = −2.755	t = 0.401	t = 1.517	t = 1.576
	p = .006	p = .689	p = .130	p = .116
Constant	−0.001	1.823***	1.227***	1.226***
	(0.056)	(0.020)	(0.009)	(0.008)
	t = −0.017	t = 93.322	t = 142.974	t = 148.962
	p = .987	p = .000	p = .000	p = .000
Log likelihood	−1,045.002	−250.443	718.707	724.402
Akaike information criterion	2,112.003	522.887	−1,415.414	−1,426.803
Bayesian information criterion	2,165.867	576.993	−1,361.082	−1,372.472

Notes: Table depicts estimate coefficients, standard error, t-value and p value.

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

Table 4.

Summary of Linear Mixed Model for Attribution of Responsibility.

	Dependent Variable
	Doctors	Patients	Family Caregivers	God	Chance/Luck
	(1)	(2)	(3)	(4)	(5)
Decision-making	0.012	−0.024	−0.045	0.001	0.008*
	(0.007)	(0.019)	(0.026)	(0.007)	(0.004)
	t = 1.605	t = −1.247	t = −1.723	t = 0.145	t = 1.965
	p = .109	p = .213	p = .085	p = .885	p = .050
Financial status	−0.017*	0.039**	0.048*	−0.021**	−0.003
	(0.007)	(0.014)	(0.021)	(0.007)	(0.004)
	t = −2.332	t = 2.776	t = 2.357	t = −2.960	t = −0.693
	p = .020	p = .006	p = .019	p = .004	p = .489
Decision-making: Financial status	0.0002	−0.016	0.006	0.0004	−0.0002
	(0.006)	(0.013)	(0.018)	(0.006)	(0.004)
	t = 0.037	t = −1.317	t = 0.324	t = 0.071	t = −0.053
	p = .971	p = .188	p = .747	p = .944	p = .958
Constant	1.688***	−0.000	2.409***	1.586***	0.446***
	(0.025)	(0.056)	(0.074)	(0.027)	(0.014)
	t = 68.257	t = −0.000	t = 32.513	t = 58.723	t = 32.285
	p = .000	p = 1.000	p = .000	p = .000	p = .000
Log likelihood	−227.201	−1,060.826	−1,398.061	−280.374	328.577
Akaike information criterion	476.403	2,143.653	2,818.122	582.748	−645.154
Bayesian information criterion	530.735	2,197.985	2,872.454	637.079	−615.519

Notes: Table depicts estimate coefficients, standard error, t-value and p value.

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

Trust in Doctors

High significant main effects were found for both decision-making style (βˆ = −0.050, p = .002) and financial status (βˆ = −0.089, p = .00001) on scores of trust in doctors. The negative coefficient for decision-making style suggests that with a shift from paternalistic to a shared decision-making style, the trust in doctor tends to increase. Similarly, the negative coefficient for financial status indicates that with the decrease in financial status, that is, from high to low, trust in doctors tends to increase. A significant interaction effect for decision-making style and financial status was also found (βˆ = −0.037, p = .006). This means that the impact of both decision-making and financial status is modified by the presence of the other variable.

Further, post-hoc analysis suggests that the scores for trust in doctors in HP condition are significantly different from HS condition (t-value = −4.169, p = .0002), LS condition (t-value= −5.386, p <.00001) and LP condition (t-value = −5.583, p <.00001). This can be more clearly seen in Figure 1 which depicts the mean scores of trust in doctors when grouped by decision-making and financial status. The graph shows that overall trust in doctors is lowest for HP condition. The conditions for low financial status elicited higher trust in doctors’ scores as compared to conditions with high financial status. Similarly, conditions with shared decision-making elicited higher trust in doctors scores as compared to paternalistic decision-making conditions. The pseudo-R² for the entire model was 0.825, and for fixed effects was 0.0115. It means that though the model is quite robust as it accounts for 82.5% of variation in scores for trust in doctors, only 1.15% of this variance is attributed to fixed effects.

Figure 1.

Mean Scores of Trust in Doctors: Grouped by Decision-making and Financial Status.

Patient’s Agency

High significant main effects were found for both decision-making style (βˆ = −0.018, p = .021) and financial status (βˆ = 0.071, p = .000) for scores on perceived patient’s agency. The negative coefficient for decision-making style suggests that with a shift from paternalistic to a shared decision-making style, the patient’s agency tends to increase. The positive coefficient for financial status indicates that with the decrease in financial status, that is, from high to low, perceived patient’s agency also tends to decrease. No significant interaction effect for decision-making style and financial status was found (βˆ = 0.002, p = .689).

Further, post-hoc analysis suggests that the scores for patient’s agency were significantly different in four groups. Specifically, differences between: HP condition and LP condition (t-value= 6.482, p <.0001), HP condition and LS condition (t-value= 4.489, p = .0001), HS condition and LP condition (t-value= 6.975, p <.0001) and HS condition and LS condition (t-value= 6.059, p <.0001). Figure 2 depicts the mean scores of perceived patient’s agency when grouped by decision-making and financial status. The graph shows that patient’s agency was perceived to be much higher in conditions of high financial status as compared to low financial status. Further, patient’s agency was perceived to be lower for paternalistic decision conditions as compared to shared decision conditions. The pseudo-R² for the entire model was 0.749, and for fixed effects was 0.0356. It means that though the model is quite robust as it accounts for 74.9% of variation in scores for patient’s agency, only 3.56% of this variance is attributed to fixed effects.

Figure 2.

Mean Scores of Patient’s Agency: Grouped by Decision-making and Financial Status.

Emotional Distress

Patients

High significant main effects were found for both decision-making style (βˆ = 0.015, p = .00001) and financial status (βˆ = −0.020, p = .000) for scores of perceived emotional distress for patients. The main effect for decision-making style suggests that emotional distress for patients is higher in paternalistic decision condition as compared to shared decision condition. The negative coefficient for financial status suggests that with the decrease in financial status, that is, from high to low, perceived emotional distress for patients tends to increase. No significant interaction effect for decision-making style and financial status was found (βˆ = 0.004, p = .130).

Post-hoc analysis suggests that apart from the difference between HP condition and LS condition, all other groups had significant differences. Specifically, significant differences were found between HP and HS condition (t-value= 4.744, p <.0001), HP and LP condition (t-value= −4.461, p = .0001), HS and LP condition (t-value= −8.067, p <.0001), HS and LS condition (t-value= −6.360, p <.0001) and LP and LS condition (t-value= 2.975, p = .0162). Figure 3 depicts the mean scores of perceived emotional distress for patients when grouped by decision-making and financial status. The graph shows that emotional distress for patients was perceived to be much higher in conditions of low financial status as compared to high financial status. Similarly, emotional distress for patients was perceived to be higher for paternalistic decision conditions as compared to shared decision conditions. The pseudo-R² for the entire model was 0.784, and for fixed effects was 0.0253. It means that though the model is quite robust as it accounts for 78.4% of variation in scores for emotional distress for patients, only 2.53% of this variance is attributed to fixed effects.

Figure 3.

Mean Scores of Emotional Distress for Patients: Grouped by Decision-making and Financial Status.

Family Caregivers

High significant main effects were found for both decision-making style (βˆ = 0.016, p = .0000) and financial status (βˆ = −0.021, p = .000) for scores of perceived emotional distress for family caregivers. The main effect for decision-making style suggests that emotional distress for family caregivers is higher in paternalistic decision condition as compared to shared decision condition. The negative coefficient for the main effect of financial status suggests that with the decrease in financial status, that is, from high to low, perceived emotional distress for family caregivers tends to increase. No significant interaction effect for decision-making style and financial status was found (βˆ = 0.004, p = .116).

Post-hoc analysis suggests that apart from the difference between HP condition and LS condition, all other groups had significant differences. Specifically, significant differences were found between HP and HS condition (t-value=5.140, p <.0001), HP and LP condition (t-value= −4.309, p = .0001), HS and LP condition (t-value= −8.244, p <.0001), HS and LS condition (t-value= −6.299, p <.0001) and LP and LS condition (t-value= 3.132, p = .0099). Figure 4 depicts the mean scores of perceived emotional distress for family caregivers when grouped by decision-making and financial status. The graph shows that emotional distress for family caregivers was perceived to be much higher in conditions of low financial status as compared to high financial status. Similarly, emotional distress for family caregivers was perceived to be higher for paternalistic decision conditions as compared to shared decision conditions. The pseudo-R² for the entire model was 0.739, and for fixed effects was 0.0291. It means that though the model is quite robust as it accounts for 73.9% of variation in scores for emotional distress for family caregivers, only 2.91% of this variance is attributed to the fixed effects.

Figure 4.

Mean Scores of Emotional Distress for Family Caregiver: Grouped by Decision-making and Financial Status.

Attribution of Responsibility

Table 4 depicts the results for LMMs for the five factors of attribution of responsibility for taking the decision in case the treatment hampered the patient’s life or is unable to eliminate cancer. For all the factors, except for the factor of chance or luck, a significant main effect for financial status was found (see Table 4). For the factors of doctor and god, the negative estimates reflect that as the financial status decreases from high to low, the attribution of responsibility to these factors increases. Conversely, for the factors of Patients and Family caregivers, as the financial status decreased from high to low, the attribution of responsibility to these factors also decreased. Apart from the factor of Chance, no significant main effect was found for decision-making (βˆ = 0.008, p = .050). This suggests that attribution of responsibility for only the factor of chance was higher in conditions of paternalistic decision as compared to shared decision. For all the other factors, there were no significant differences in attribution of responsibility due to decision-making conditions. No significant interaction effects were found for all five factors (see Table 4).

Post-hoc analysis suggests significant differences only between the HS and LP condition for attribution of responsibility to doctors (t-value= −2.589, p = .0496), patients (t-value= 2.604, p = .0477) and family caregivers (t-value= 2.976, p = .0168). No post-hoc differences were found for the factors of god and chance. Surprisingly, the pseudo-R² for the fixed effects for all five factors were exceptionally low: doctors (0.22%), patients (0.24%), family caregivers (0.24%), god (0.23%) and chance (0.11%). This suggests that though the main effects are significant, they barely account for the variation in the scores of attribution to responsibility. However, Figure 5 helps in understanding the results of the attribution of responsibility scores across all five factors. It depicts that the highest attribution of responsibility is given to Chance, followed by doctors and god, whereas the lowest attribution of responsibility is given to patients. In condition of high financial status, patients and family caregivers are attributed more responsibility as compared to low financial status condition.

Figure 5.

Mean Scores of Attribution of Responsibility: Grouped by Decision-making and Financial Status.

Discussion

The study aimed to explore the influence of individual (paternalistic and shared decision-making style) and contextual (financial status of the patients) factors on certain aspects of decision process, namely, trust in doctors, patient’s agency, emotional distress, and attribution of responsibility.

Influence of Paternalistic Versus Shared Decision-making on Decision Process

The findings support H₁, H₂ and H₃, indicating that shared decision-making correlates with higher trust on doctors, higher patients’ agency and lower emotional distress for both patients and family caregivers when compared to paternalistic decision-making. Contrarily, the study does not support H₄. While a lower attribution of responsibility to chance/luck was found for shared decision-making compared to paternalistic decision-making, no such significant differences were found for other factors, namely, doctors, God, patients and family caregivers.

Previous research on shared decision-making yields conflicting evidence regarding its potential benefits, with some studies suggesting positive associations with patient satisfaction and health outcomes (Bomhof-Roordink et al., 2019; Entwistle, 2004; Entwistle et al., 2012; Josfeld et al., 2021), while others express scepticism about its practical application (Gillman & Ferrer, 2021; Guadalajara et al., 2022; Ozdemir & Finkelstein, 2018). In this study, experimental conditions of shared decision-making were found to be linked to heightened trust, reduced emotional distress, and increased patient agency. A recent mixed-method study involving 369 surrogate family decision-makers identified five shared decision-making behaviours—‘discussing surrogate’s role, explaining medical condition and prognosis, providing emotional support, assessing understanding, and eliciting patient’s values and preferences’ (Lincoln et al., 2023, p. 1214). Each of these elements was associated with a 0.37-point increase in trust in the treating physician (Lincoln et al., 2023).

Regardless of contextual factors, the findings indicate that patient’s agency was consistently low for conditions of paternalistic decision-making, wherein doctors solely bear decision responsibilities (Davies & Parker, 2022). In a study, Blok et al. (2023) interviewed 40 family caregivers of patients undergoing life-assistive treatment. They found that ‘anxious family caregivers’ expressed distress due to ‘lack of information about health care system processes’ (Blok et al., 2023, p. 25). Similarly, those in borderline anxiety group reported that ‘not understanding the reasons for the medical team’s care decisions contributed to their distress’ (Blok et al., 2023, p. 26). Such studies suggest that since shared decision-making promotes open communication (Guadalajara et al., 2022), there is an increased patients’ and family caregivers’ participation in the treatment process (Lincoln et al., 2023) and lower treatment anxiety (Blok et al., 2023).

Surprisingly, there were no significant differences between the two decision-making styles in the attribution of responsibility for treatment decisions, except in case of chance/luck. Though paternalistic decision-making is associated with increased responsibility on the part of doctors compared to patients or family caregivers (Davies & Parker, 2022), in this study no such distinctions were found. Further, previous literature suggests the tendency of patients and family caregivers to relegate to metaphysical beliefs, for instance, beliefs in god, karma or chance/luck to explain the occurrence of illness (Kohli & Dalal, 2016). The current findings also indicate that specifically in paternalistic decision-making conditions, the attribution of decision responsibility was higher for the factor of chance/luck. This could so as the patients feel less in control of the treatment process; they tend to tilt more towards external locus of control for the treatment decisions (Kohli & Dalal, 2016; White et al., 2021).

Role of Low Versus High Financial Status in Decision-making

Financial toxicity remains a nascent area of investigation within the literature, with most studies primarily focusing on its prevalence (Abrams et al., 2021; Boby et al., 2021; Prinja et al., 2023), identifying high-risk groups (Boby et al., 2021; Lentz et al., 2019; Prinja et al., 2023), and establishing its connections with quality of life, well-being and physical health (Abrams et al., 2021; Carrera et al., 2018; Zafar & Peppercorn, 2023). However, its impact on decision-making in cancer treatment remains an understudied area. This study contributes to the evolving body of literature by elucidating the associations between financial status and crucial aspects of the decision-making process.

The findings of the study support H₅, H₆ and H₇, indicating that conditions of low financial status elicited higher trust in doctors, lower patient’s agency and high emotional distress for both patients and family caregivers as compared to high financial status conditions. Previous studies link low financial status with lower awareness of cancer and treatment therapies and lower access to informational resources (Sahu et al., 2020; Singh et al., 2016). Moreover, studies on paternalistic decision-making have linked lower financial status with higher trust on doctors (Driever et al., 2021; Jacob, 2014; Ram Prakash & Lingam, 2021). Research indicates that patients from low financial status are often not able to complete treatment, lose their daily wages, go under tremendous financial debt, are not able to take second opinions and thus have lesser options as compared to patients with financial security (Abrams et al., 2021; Boby et al., 2021; Zafar & Peppercorn, 2023). In a qualitative study exploring gender barriers in utilisation of government insurance schemes, Ram Prakash and Lingam (2021) found that female patients’ financial dependence and lack of control over resources severely affected their health-related decision-making. They were mostly dependent on male members of the family for financing even outpatient consultations. Further, for male patients belonging to ‘very poor families’, treatment decisions such as hospitalisation and surgery were often delayed due to concerns about the potential loss of daily wages during both surgical and aftercare period (Ram Prakash & Lingam, 2021). Such studies indicate how crucial treatment-related decisions are inherently linked to the financial status of the patients. However, there are barely any studies that empirically illuminate the impact of the financial status of patients on the decision process in cancer treatment.

Further, H₈ was partially supported. The findings indicate that attribution of decision responsibility in low financial status conditions was found to be higher for doctors and god, and conversely, lower for patients and family caregivers, as compared to the high financial status group. However, there was no significant difference between the two groups for the factor of chance or luck. Factors such as low knowledge and awareness among patients belonging to low financial status (Sahu et al., 2020; Singh et al., 2016), reduced agency in choosing healthcare facilities (Ram Prakash & Lingam, 2021) and a high reliance on doctors’ authority (Bhakuni, 2020; Subramani, 2019) could probably explain such an external locus of control—wherein god and doctors are attributed higher decision responsibility than patients and family caregivers.

Interaction of Individual and Contextual Variables

The study did not find enough evidence to support H₉, such that, the interaction of decision-making style and financial status was found to be significant only for trust in doctors. For all other variables, patient’s agency emotional distress for patients and family caregivers, attribution of responsibility (doctors, patients, family caregivers, god and chance/luck), no significant interaction effect was found. Interestingly, the findings indicate that trust in doctors was lowest for the interaction of high financial status and paternalistic decision style, as compared to all other groups. In a study involving 606 patients, Kraetschmer et al. (2004) found that patients who preferred an autonomous role in decision-making exhibited relatively low levels of trust in physicians. Alternatively, those who preferred a shared role had moderate levels of trust, while those who preferred a passive role were more likely to have blind trust.

Bhakuni (2020) makes a cogent argument that the Supreme Court of India has cited poverty and trust in doctors as reasons to adopt a paternalistic view of informed consent, such that the doctors determine the amount of information to be given to the patient. Further, researchers like Gopichandran (2019), Subramani (2017, 2019), and Bhakuni (2020) have also commented upon the role of trust in defining the boundaries of doctor–patient relationship, emphasising how such trust is often at intersection of prevailing sociolegal conditions and the various social identities of the patient.

God, Fate, and Fatalism

Interestingly, the findings indicate pervasively high levels of attribution of decision responsibility to chance/luck and god, regardless of financial status or decision-making style. Previous studies suggest that belief in god, fatalism and doctrine of karma served as causal explanations for why the illness occurred, providing meaning to the patients’ suffering and facilitating acceptance of the reality of cancer (Kohli & Dalal, 2016; Sprik et al., 2021; Vardar et al., 2021).

There is scant literature that explores the role of attribution of decision responsibility in cancer treatment to god or chance/luck (Dalal & Misra, 2006; Kohli & Dalal, 2016). While most studies link belief in god or fatalism to positive physical outcomes (Sprik et al., 2021; Vardar et al., 2021), functional well-being (Leyva et al., 2014), sensemaking (Keeley et al., 2009), and positive coping (Vardar et al., 2021), few have investigated the role of belief in god and/or fate in cancer-related decision-making. For instance, Dopelt et al. (2022) using qualitative interviews explored cancer patients’ and family caregivers’ health locus of control with respect to clinical decision-making. They found that some patients relied on the belief in fate, leading them to adopt a passive stance in decision-making and refrain from asking questions. Few other researchers have explored cultural beliefs, such as fate, god and karma in relation to the attribution of cause of illness (Agrawal & Dalal, 1993; Dalal, 2000; Dalal & Misra, 2006; Kohli & Dalal, 2016). This study contributes to the limited literature as it investigates the extent of individuals’ attribution of responsibility to cultural metaphysical beliefs in cancer treatment decision-making.

Conclusion

Overall, this study deepens the understanding of how the individualistic decision-making styles (paternalistic versus shared) are linked to various crucial aspect of the decision process, particularly within the context of patients’ financial status (low versus high). Though shared decision-making is often characterised as a utopian concept (Gillman & Ferrer, 2021; Ozdemir & Finkelstein, 2018), the results indicate that a shift towards shared approaches does yield qualitatively superior outcomes in terms of increased trust in doctors and patients’ agency along with reduced emotional distress. The study also emphasises the role of cultural metaphysical beliefs in attributing responsibility for treatment decisions.

Further, it is crucial to acknowledge that the decision-making style is influenced by the contextual constraints of the patients. Given that a significant proportion of cancer patients in India come from financially disadvantaged backgrounds (Sathishkumar et al., 2022), this study makes substantial contributions by revealing that low financial status is associated with higher trust in doctors, lower patient agency and increased emotional distress for both patients and family caregivers. The move towards shared decision-making, considering the contextual realities of patients, could be instrumental in addressing critical issues, such as prevalence of collusion, superficial role of informed consent and high reliance on doctors’ authority or expertise. The shared decision-making approach promotes open communication between doctors, patients and family caregivers, offering a potential solution to mitigate these challenges.

Footnotes

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.

ORCID iD

Nishtha Jain

References

Abrams

H. R.

, Durbin

, Huang

C. X.

, Johnson

S. F.

, Nayak

R. K.

, Zahner

G. J.

, & Peppercorn

(2021). Financial toxicity in cancer care: Origins, impact, and solutions. Translational Behavioral Medicine , 11(11), 2043–2054. https://doi.org/10.1093/tbm/ibab091

Agrawal

, & Dalal

A. K.

(1993). Beliefs about the world and recovery from myocardial infarction. Journal of Social Psychology , 133(3), 385–394. https://doi.org/10.1080/00224545.1993.9712157

Bates

, Maechler

, Bolker

, & Walker

(2023). lme4: Linear mixed-effects models using Eigen and S4 . https://github.com/lme4/lme4/

Betsch

, & Glöckner

(2010). Intuition in judgment and decision making: Extensive thinking without effort. Psychological Inquiry , 21(4), 279–294. https://doi.org/10.1080/1047840X.2010.517737

Bhakuni

(2020). Informed consent to clinical research in India: A private law remedy. Medical Law International , 20(3), 256–283. https://doi.org/10.1177/0968533220958185

Blok

A. C.

, Valley

T. S.

, Weston

L. E.

, Miller

, Lipman

, & Krein

S. L.

(2023). Factors affecting psychological distress in family caregivers of critically ill patients: A qualitative study. American Journal of Critical Care , 32(1), 21–30. https://doi.org/10.4037/ajcc2023593

Boby

J. M.

, Rajappa

, & Mathew

(2021). Financial toxicity in cancer care in India: A systematic review. The Lancet Oncology , 22(12), e541–e549. https://doi.org/10.1016/S1470-2045(21)00468-X

Bomhof-Roordink

, Fischer

M. J.

, van Duijn-Bakker

, Baas-Thijssen

M. C.

, van der Weijden

, Stiggelbout

A. M.

, & Pieterse

A. H.

(2019). Shared decision making in oncology: A model based on patients’, health care professionals’, and researchers’ views. Psycho-Oncology , 28(1), 139–146. https://doi.org/10.1002/PON.4923

Carrera

P. M.

, Kantarjian

H. M.

, & Blinder

V. S.

(2018). The financial burden and distress of patients with cancer: Understanding and stepping-up action on the financial toxicity of cancer treatment. CA Cancer Journal for Clinicians , 68(2), 165. https://doi.org/10.3322/CAAC.21443

10.

Chandra

P. S.

, Chaturvedi

S. K.

, Kumar

, Subbakrishna

D. K.

, Channabasavanna

S. M.

, & Anantha

(1998). Awareness of diagnosis and psychiatric morbidity among cancer patients—A study from South India. Journal of Psychosomatic Research , 45(3), 257–261. https://doi.org/10.1016/S0022-3999(98)00012-9

11.

Chaturvedi

, Loiselle

, & Chandra

(2009). Communication with relatives and collusion in palliative care: A cross-cultural perspective. Indian Journal of Palliative Care , 15(1), 9. https://doi.org/10.4103/0973-1075.53485

12.

Chittem

, Maya

, & Chawak

(2021). Nondisclosure of a cancer diagnosis and prognosis: Recommendations for future research and practice. Indian Journal of Cancer , 58(2), 164. https://doi.org/10.4103/IJC.IJC_740_19

13.

Dalal

A. K.

(2000). Living with a chronic disease: Healing and psychological adjustment in Indian society. Psychology and Developing Societies , 12(1), 67–81. https://doi.org/10.1177/097133360001200105

14.

Dalal

A. K.

, & Misra

(2006). Psychology of health and well-being: Some emerging perspectives. Psychological Studies , 51(2–3), 91–104.

15.

Datta

S. S.

, Tripathi

, Varghese

, Logan

, Gessler

, Chatterjee

, Bhaumik

, & Menon

(2017). Pivotal role of families in doctor–patient communication in oncology: A qualitative study of patients, their relatives and cancer clinicians. European Journal of Cancer Care , 26(5), e12543. https://doi.org/10.1111/ECC.12543

16.

Davies

, & Parker

(2022). Doctors as appointed fiduciaries: A supplemental model for medical decision-making. Cambridge Quarterly of Healthcare Ethics , 31(1), 23–33. https://doi.org/10.1017/S096318012100044X

17.

Dopelt

, Bashkin

, Asna

, & Davidovitch

(2022). Health locus of control in cancer patient and oncologist decision-making: An exploratory qualitative study. PLoS One , 17(1), e0263086. https://doi.org/10.1371/journal.pone.0263086

18.

Driever

E. M.

, Stiggelbout

A. M.

, & Brand

P. L. P.

(2020). Shared decision making: Physicians’ preferred role, usual role and their perception of its key components. Patient Education and Counseling , 103(1), 77–82. https://doi.org/10.1016/J.PEC.2019.08.004

19.

Driever

E. M.

, Tolhuizen

I. M.

, Duvivier

R. J.

, Stiggelbout

A. M.

, & Brand

P. L. P.

(2021). Why do medical residents prefer paternalistic decision making? An interview study. BMC Medical Education , 22(1), 1–10. https://doi.org/10.1186/s12909-022-03203-2

20.

Elkin

E. B.

, Kim

S. H. M.

, Casper

E. S.

, Kissane

D. W.

, & Schrag

(2007). Desire for information and involvement in treatment decisions: Elderly cancer patients’ preferences and their physicians’ perceptions. Journal of Clinical Oncology , 25(33), 5275–5280. https://doi.org/10.1200/JCO.2007.11.1922

21.

Entwistle

(2004). Trust and shared decision-making: An emerging research agenda. Health Expectations , 7(4), 273. https://doi.org/10.1111/J.1369-7625.2004.00304.X

22.

Entwistle

V. A.

, Cribb

, & Watt

I. S.

(2012). Shared decision-making: Enhancing the clinical relevance. Journal of the Royal Society of Medicine , 105(10), 416–421. https://doi.org/10.1258/JRSM.2012.120039/FORMAT/EPUB

23.

Fox

, Weisberg

, & Price

(2023). CAR: Companion to applied regression . https://r-forge.r-project.org/projects/car/

24.

Gigerenzer

, & Gaissmaier

(2011). Heuristic decision making. Annual Review of Psychology , 62(1), 451–482. https://doi.org/10.1146/ANNUREV-PSYCH-120709-145346

25.

Gillman

A. S.

, & Ferrer

R. A.

(2021). Opportunities for theory-informed decision science in cancer control. Translational Behavioral Medicine , 11(11), 2055–2064. https://doi.org/10.1093/tbm/ibab141

26.

Gopichandran

(2019). Dynamics of trust in doctor–patient relationship in India . Springer. https://doi.org/10.1007/978-981-15-0346-7

27.

Guadalajara

, Lopez-Fernandez

, León Arellano

, Domínguez-Prieto

, Caramés

, & Garcia-Olmo

(2022). The role of shared decision-making in personalised medicine: Opening the debate. Pharmaceuticals , 15(2), 215. https://doi.org/10.3390/PH15020215

28.

F. J.

, Anat

D. S.

, & Longnecker

(2010). Doctor–patient communication: A review. The Ochsner Journal , 10(1), 38–43. https://www.ochsnerjournal.org/content/ochjnl/10/1/38.full.pdf

29.

Hahlweg

, Härter

, Nestoriuc

, & Scholl

(2017). How are decisions made in cancer care? A qualitative study using participant observation of current practice. BMJ Open , 7(9), e016360. https://doi.org/10.1136/bmjopen-2017-016360

30.

Hall

M. A.

(2006). Researching medical trust in the United States. Journal of Health Organisation and Management , 20(5), 456–467. https://doi.org/10.1108/14777260610701812

31.

Harding

, Salins

, Sharan

, & Ekstrand

M. L.

(2022). Health literacy in communication, decision-making and outcomes among cancer patients, their families and clinicians in India: A multicentre cross-sectional qualitative study. Psycho-Oncology , 31(3), 532–540. https://doi.org/10.1002/PON.5838

32.

Heidenreich

, Bremer

, Materstvedt

L. J.

, Tidefelt

, & Svantesson

(2017). Relational autonomy in the care of the vulnerable: Health care professionals’ reasoning in Moral Case Deliberation (MCD). Medicine, Health Care and Philosophy , 21(4), 1–11. https://doi.org/10.1007/s11019-017-9818-6

33.

Howell

D. C.

(2019). Statistical methods for psychology (8th ed.). Thomson Wadsworth.

34.

Jacob

K. S.

(2014). Informed consent and India. The National Medical Journal of India , 27(1), 35–38.

35.

Josfeld

, Keinki

, Pammer

, Zomorodbakhsch

, & Hübner

(2021). Cancer patients’ perspective on shared decision-making and decision aids in oncology. Journal of Cancer Research and Clinical Oncology , 147(6), 1725–1732. https://doi.org/10.1007/S00432-021-03579-6/FIGURES/4

36.

Keeley

, Wright

, & Condit

C. M.

(2009). Functions of health fatalism: Fatalistic talk as face saving, uncertainty management, stress relief and sense making. Sociology of Health & Illness , 31(5), 734–747. https://doi.org/10.1111/J.1467-9566.2009.01164.X

37.

Klein

(2017). Sources of power: How people make decisions. Sources of power (20th ed.). The MIT Press. https://doi.org/10.7551/MITPRESS/11307.001.0001

38.

Kohli

, & Dalal

A. K.

(2016). Culture as a factor in causal understanding of illness : A study of cancer patients. Psychology and Developing Societies , 10(2), 115–129. https://doi.org/10.1177/097133369801000202

39.

Kraetschmer

, Sharpe

, Urowitz

, & Deber

R. B.

(2004). How does trust affect patient preferences for participation in decision-making? Health Expectations , 7(4), 326. https://doi.org/10.1111/J.1369-7625.2004.00296.X

40.

Lantos

J. D.

(2021). Introduction: The fascinating synergy of shared decision making. In Lantos

J. D.

(Ed.), The ethics of shared decision making (pp. 1–6). Oxford Academic. https://doi.org/10.1093/med/9780197598573.003.0001

41.

Lazcano-Ponce

, Angeles-Llerenas

, Rodríguez-Valentín

, Salvador-Carulla

, Domínguez-Esponda

, Astudillo-García

C. I.

, Madrigal-de León

, & Katz

(2020). Communication patterns in the doctor–patient relationship: evaluating determinants associated with low paternalism in Mexico. BMC Medical Ethics , 21(1), 1–11. https://doi.org/10.1186/S12910-020-00566-3/TABLES/3

42.

Lenth

R. V.

(2023). EMMEANS: Estimated marginal means, aka least-squares means . https://github.com/rvlenth/emmeans

43.

Lentz

, Benson

A. B.

, & Kircher

(2019). Financial toxicity in cancer care: Prevalence, causes, consequences, and reduction strategies. Journal of Surgical Oncology , 120(1), 85–92. https://doi.org/10.1002/JSO.25374

44.

Leyva

, Allen

J. D.

, Tom

L. S.

, Ospino

, Torres

M. I.

, & Abraido-Lanza

A. F.

(2014). Religion, fatalism, and cancer control: A qualitative study among Hispanic Catholics. American Journal of Health Behavior , 38(6), 849. https://doi.org/10.5993/AJHB.38.6.6

45.

Lincoln

T. E.

, Buddadhumaruk

, Arnold

R. M.

, Scheunemann

L. P.

, Ernecoff

N. C.

, Chang

C. C. H.

, Carson

S. S.

, Hough

C. L.

, Curtis

J. R.

, Anderson

, Steingrub

, Peterson

M. W.

, Lo

, Matthay

M. A.

, & White

D. B.

(2023). Association between shared decision-making during family meetings and surrogates’ trust in their ICU physician. Chest , 163(5), 1214–1224. https://doi.org/10.1016/J.CHEST.2022.10.028

46.

Lüdecke

, Ben-Shachar

, Patil

, Waggoner

, & Makowski

(2021). Performance: An R package for assessment, comparison and testing of statistical models. Journal of Open Source Software , 6(60), 3139. https://doi.org/10.21105/joss.03139

47.

Mathew

, George

P. S.

, Kunnambath

, Mathew

B. S.

, Kumar

, Syampramod

, & Booth

C. M.

(2020). Educational status, cancer stage, and survival in South India: A population-based study. JCO Global Oncology , 6(6), 1704–1711. https://doi.org/10.1200/GO.20.00259

48.

Mathew

, Mohanti

B. K.

, Tewari

, Kabra

, Gulia

, Bajpai

, Munshi

, Mathew

, Mohanti

B. K.

, Tewari

, Kabra

, Gulia

, Bajpai

, & Munshi

(2021). Collusion: The facade and its implications on total pain management in palliative care. Indian Journal of Palliative Care , 27(1), 176–179. https://doi.org/10.4103/IJPC.IJPC_81_20

49.

Nakagawa

, Johnson

P. C. D.

, & Schielzeth

(2017). The coefficient of determination R² and intra-class correlation coefficient from generalized linear mixed-effects models revisited and expanded. Journal of the Royal Society Interface , 14(134), 20170213. https://doi.org/10.1098/RSIF.2017.0213

50.

O’Hair

, Villagran

M. M.

, Wittenberg

, Brown

, Ferguson

, Hall

H. T.

, & Doty

(2003). Cancer survivorship and agency model: Implications for patient choice, decision making, and influence. Health Communication , 15(2), 193–202. https://doi.org/10.1207/S15327027HC1502_7

51.

Oprea

, Ardito

, & Ciani

(2023). Implementing shared decision-making interventions in breast cancer clinical practice: A scoping review. BMC Medical Informatics and Decision Making , 23(1). https://doi.org/10.1186/s12911-023-02263-8

52.

Ozdemir

, & Finkelstein

E. A.

(2018). Cognitive bias: The downside of shared decision making. JCO Clinical Cancer Informatics , 2, 1–10. https://doi.org/10.1200/CCI.18.00011

53.

Panda

(2023). Public trust in government doctors and hospitals in India. International Journal of Social Economics , 50(11), 1602–1617. https://doi.org/10.1108/IJSE-07-2022-0498/FULL/PDF

54.

Papathanassoglou

E. D. E.

, Karanikola

M. N. K.

, Kalafati

, Giannakopoulou

, Lemonidou

, & Albarran

J. W.

(2012). Professional autonomy, collaboration with physicians, and moral distress among European intensive care nurses. American Journal of Critical Care , 21(2), e41–e52. https://doi.org/10.4037/ajcc2012205

55.

Pollard

, Bansback

, & Bryan

(2015). Physician attitudes toward shared decision making: A systematic review. Patient Education and Counseling , 98(9), 1046–1057. https://doi.org/10.1016/J.PEC.2015.05.004

56.

Prinja

, Dixit

, Gupta

, Dhankhar

, Kataki

A. C.

, Roy

P. S.

, Mehra

, Kumar

, Singh

, Malhotra

, Goyal

, Rajsekar

, Krishnamurthy

M. N.

, & Gupta

(2023). Financial toxicity of cancer treatment in India: Towards closing the cancer care gap. Frontiers in Public Health , 11. https://doi.org/10.3389/fpubh.2023.1065737

57.

Ram Prakash

, & Lingam

(2021). Why is women’s utilization of a publicly funded health insurance low?: A qualitative study in Tamil Nadu, India. BMC Public Health , 21(1), 1–21. https://doi.org/10.1186/S12889-021-10352-4/TABLES/4

58.

RStudio Team . (2022). RStudio: Integrated development environment for R (2022.7.2.576). RStudio, PBC. http://www.rstudio.com/

59.

Sahu

D. P.

, Subba

S. H.

, & Giri

P. P.

(2020). Cancer awareness and attitude towards cancer screening in India: A narrative review. Journal of Family Medicine and Primary Care , 9(5), 2218. https://doi.org/10.4103/JFMPC.JFMPC_145_20

60.

Sankar

S. D.

, Dhanapal

, Krishnaraj

, Karra

, Natesan

, & Gomathi Shankar

(2018). Desire for information and preference for participation in treatment decisions in patients with cancer presenting to the department of general surgery in a tertiary care hospital in India. Journal of Global Oncology , 4(4), 1–10. https://doi.org/10.1200/JGO.17.00144

61.

Sareen

, & Dutt

(2019). Informed consent in medical decision making in India. Journal of Counselling and Family Therapy , 1(1), 30–37. https://doi.org/http://dx.doi.org/10.5281/zenodo.2597154

62.

Sathishkumar

, Chaturvedi

, Das

, Stephen

, & Mathur

(2022). Cancer incidence estimates for 2022 & projection for 2025: Result from National Cancer Registry Programme, India. The Indian Journal of Medical Research , 156(4 & 5), 598–607. https://doi.org/10.4103/IJMR.IJMR_1821_22

63.

Singh

, Sharma

, Bhandari

, & Kaur

(2016). Knowledge, awareness and practice of ethics among doctors in tertiary care hospital. Indian Journal of Pharmacology , 48(7), 89–93. https://doi.org/10.4103/0253-7613.193320

64.

Singmann

, & Kellen

(2019). An Introduction to mixed models for experimental psychology. In D. H. Spieler & E. Schumacher (Eds), New methods in cognitive psychology (pp. 4–31). Psychology Press. https://cran.r-project.org/web/packages/afex/vignettes/introduction-mixed-models.pdf

65.

Sparks

C. S.

(2018). Basic hierarchical models—Random intercepts and slopes . RPubs by RStudio. https://rpubs.com/corey_sparks/420770

66.

Sprik

P. J.

, Tata

, Kelly

, & Fitchett

(2021). Religious/Spiritual concerns of patients with brain cancer and their caregivers. Annals of Palliative Medicine , 10(1). https://doi.org/10.21037/apm-20-813

67.

Subramani

(2017). Patient autonomy within real or valid consent: Samira Kohli’s case. Indian Journal of Medical Ethics , 2(3), 184–189. https://doi.org/10.20529/IJME.2017.038

68.

Subramani

(2019). The rhetoric of the ‘passive patient’ in Indian medical negligence cases. Asian Bioethics Review , 11(4), 366. https://doi.org/10.1007/S41649-019-00106-1

69.

Sutar

, Chaudhary

, & Yadav

(2022). Prevalence of collusion in cancer communications: A meta-analysis. Psycho-Oncology , 31(3), 372–387. https://doi.org/10.1002/PON.5824

70.

Thompson

G. A.

, & Whiffen

L. H.

(2018). Can physicians demonstrate high quality care using paternalistic practices? A case study of paternalism in Latino Physician–Patient interactions. Qualitative Health Research , 28(12), 1910–1922. https://doi.org/10.1177/1049732318783696

71.

Vardar

, Serçekus

, & Özkan

(2021). Spirituality and religion on coping with cancer for patients and caregivers. Journal of Education and Research in Nursing , 18(4), 462–466. https://doi.org/10.5152/jern.2021.35651

72.

Victor

, George

C. E.

, Inbaraj

, & Norman

(2018). Benefit or harm? A study on impact of collusion on the quality of life among palliative care patients. Indian Journal of Palliative Care , 24(1), 66. https://doi.org/10.4103/IJPC.IJPC_111_17

73.

Walker

E. A.

, Redfern

, & Oleson

J. J.

(2019). Linear mixed-model analysis to examine longitudinal trajectories in vocabulary depth and breadth in children who are hard of hearing. Journal of Speech, Language, and Hearing Research , 62(3), 542. https://doi.org/10.1044/2018_JSLHR-L-ASTM-18-0250

74.

Weiss

D. J.

, & Shanteau

(2021). The futility of decision making research. Studies in History and Philosophy of Science Part A , 90, 10–14. https://doi.org/10.1016/J.SHPSA.2021.08.018

75.

White

C. J. M.

, Willard

A. K.

, Baimel

, & Norenzayan

(2021). Cognitive pathways to belief in karma and belief in God. Cognitive Science , 45(1), e12935. https://doi.org/10.1111/COGS.12935

76.

Wickham

, Chang

, Henry

, Pedersen

T. L.

, Takahashi

, Wilke

, Woo

, Yutani

, & Dunnington

(2023). ggplot2: Create elegant data visualisations using the grammar of graphics . https://ggplot2.tidyverse.org

77.

Zafar

S. Y.

, & Peppercorn

(2023). Reflections on the financial toxicity of cancer: 10 years later. Oncologist , 28(8), 654–656. https://doi.org/10.1093/oncolo/oyad179