Impact when receiving a diagnosis: Additive and multiplicative effects between illness severity and perception of control

Abstract

Based on Lazarus’ transactional model of stress, this study examined how the information provided in a medical diagnosis (the severity of the illness, the effectiveness of the treatment and the self-efficacy to follow this treatment) are combined to generate perception of stress. Twenty-seven scenarios were presented to 152 volunteer students and their level of perceived stress was recorded. Results revealed that the lack of efficacy of the treatment raises the perception of stress proportionally to the different degrees of illness severity, but having a low perceived ability to follow the treatment triggers high levels of stress, particularly when the diagnosis is serious.

Keywords

diagnosis perceived control psychological distress self-efficacy verbal communication

Introduction

Receiving a medical diagnosis is a situation that all health service users have to face sooner or later. It is important to note that what is considered bad news depends on patient’s appraisal and elaboration of the information that he or she receives from the doctor (Lee et al., 2002). Likewise, the result of this appraisal has a huge impact on the patient’s wellbeing and psychological adjustment (Mager and Andrykowski, 2002). For this reason, in this study we set out to analyze how the key aspects in this communication interact to generate the perception of stress regarding the diagnosis received.

Given that the process of communicating negative diagnoses is regarded by doctors as inherently stressful, yet something they have to deal with quite often, it should come as no surprise that the majority of studies on communicating bad news have concentrated on ascertaining the strategies and action protocols that help them to minimize their stress, which entails improving skills at communicating with patients (Buckman, 1984; Baile et al., 2000).

More recently, interest has been sparked in finding out how patients evaluate and react to the information given to them on their health (Beadle et al., 2004; de Haes and Bensing, 2009; Randall and Wearn, 2005), given the implications this can have on the evolution of the disease. Thus, the adverse reactions described include negative emotions, anxiety, depression and stress, symptoms that in turn directly influence the patient’s adherence to therapy, along with his or her wellbeing and quality of life (Bargiel-Matusiewicz et al., 2005; Franks and Roesch, 2006; Sharf et al., 2005).

Fujimori and Uchitomi (2009) conducted a systematic review of cancer patients’ preferences regarding how they want to receive bad news. The authors classify the different preferences recorded into four components: contextual factors (such as the amount of time spent with the doctor), the way the news is communicated (such as the clarity and honesty of the information), the emotional support they receive from the doctor (such as leaving time for the patient to express his or her feelings) and a fourth component that includes the preferences on the type and amount of information shared. Regarding this last component, the patients mainly stress the importance of finding out what illness they have, their chance of a cure and the efficacy of the treatment to be followed.

The transactional model of stress developed by Lazarus and Folkman (1984; Lazarus, 1991) can explain why the information received on the implications of the illness and the control that can be exerted over it are factors that patients highlight as important in determining the level of stress experienced (Clarke and Stone, 2007; Ptacek and Eberhardt, 1996).

Thus, according to the transactional model, two kinds of appraisals regarding the information received combine to trigger stress: the primary appraisal and the secondary appraisal. The primary appraisal entails the importance or severity of the illness diagnosed, while the secondary appraisal focuses on the perception of control and of the resources available to successfully deal with the illness. This secondary appraisal consists first on the outcome expectancy, based on the patient’s knowledge about the efficacy of the treatment, as well as an appraisal focused more on personal resources, such as the self-efficacy, meaning the patient’s perceived ability to follow the medical guidelines (Bandura, 1997; Lazarus, 1991). According to this model, the stress appears when the person determines that the severity of the diagnosis is greater than the resources at their disposal (Franks and Roesch, 2006).

In addition to understanding that the stress perceived by the patient when receiving bad news will be related to the severity of the illness, the efficacy of the treatment and the self-efficacy about following the treatment, it also has to be taken into consideration how the person combines all these pieces of information. Thus, it is important to determine whether, for example, when a person believes that they feel more or less capable of following the medical guidelines, the level of stress they show rises proportionally to the different degrees of severity of the illness; or whether, to the contrary, at some level of severity the effect of having a low ability to deal with the illness on perceived stress is different than what we might expect. In the former scenario, this would indicate an additive relationship between primary and secondary appraisals, while in the latter it would indicate a multiplicative relationship. Identifying these two kinds of relationships would have implications on adjusting the message the best way possible to the patient’s needs and considering the most relevant variables when communicating the diagnosis.

For this reason, the purpose of this study is to identify what kind of combination exists among these different cognitive appraisals in order to generate a judgement on the ultimate level of perceived stress when patients receive a diagnosis.

Taking as a point of reference previous results on the combination of cognitive appraisals on perceived stress yielded in academic context and the perception of risk in athletes, we can detect several systematic patterns: the combination between any of the situational variables (in our case, the severity of the illness as primary appraisal and the efficacy of the treatment as one of the components of secondary appraisal) and personal variables (in our case, the perceived ability to follow the medical guidelines or self-efficacy as the other component of secondary appraisal) is always multiplicative (Lusar et al., 2010; Rovira et al., 2010). In contrast, the combination between the two situational variables is additive (Rovira et al., 2010). For this reason, the hypotheses on the kind of relationship we can expect to observe among the variables of severity of the illness, effectiveness of the treatment and self-efficacy, when generating the judgement on perceived stress levels, are the following:

The severity of the illness bears an additive relationship with the different levels of efficacy of the treatment. When the likelihood for the treatment to be successful is low the perception of stress regarding the diagnosis rises, regardless of the level of stress caused by the severity of the illness; likewise, the severity of the illness raises the perception of stress regardless of the level of stress triggered by the efficacy of the treatment.

The severity of the illness bears a multiplicative relationship with the self-efficacy. The perceived level of stress regarding the different degrees of severity of the illness that a diagnosis may present depends on the level of perceived ability to follow the medical guidelines observed in each case.

The self-efficacy bears a multiplicative relationship with knowledge about the efficacy of the treatment. The perceived level of stress regarding a person’s differing degrees of perceived ability to follow the medical guidelines when presented with a diagnosis depends on his or her knowledge about the efficacy of the treatment.

Method

Participants

The sample consisted of 152 university students (54% female), aged between 17 and 34 (M = 22.60; SD = 3.37), voluntary recruited among students attending psychology courses at Autonomous University of Barcelona.

Materials

To conduct this study, the theoretical and methodological framework of the Functional Theory of Cognition (Anderson, 1996) was adopted, which strives to determine how people use the information available to them to take decisions and form judgements. This theory has successfully been applied in studies in the field of health (Frileux et al., 2004; Munoz Sastre et al., 2000). According to this methodology, simulated scenarios were used which represented the kinds of information that patients may be confronted with when given a health diagnosis.

Twenty-seven scenarios were designed, the outcome of combining one of the three levels that each of the three variables manipulated can have in each scenario (severity, efficacy of the treatment and self-efficacy). The level of perceived stress when faced with each of the combinations presented was recorded.

Regarding the severity variable, the lowest level was represented by the description: ‘You have gone to the doctor to get the results of a test. The test results show an alteration related to a mild illness’. The intermediate level was represented by the description: ‘moderately serious illness’ and the highest level with ‘very serious illness’. Regarding the efficacy of the treatment, low efficacy was represented by the description: ‘Following the doctor’s instructions is successful in 20% of the cases’, while the intermediate and high levels were represented by ‘successful in 50% of the cases’ and ‘in 90% of the cases’. Finally, the lowest level of self-efficacy was represented by the description: ‘You feel unable to follow the doctor’s guidelines’, while the intermediate level stated ‘you feel moderately able’ and the high level ‘you feel very able’. An example of a scenario which corresponds to a high degree of severity, low efficacy of the treatment and average perceived ability to follow the medical guidelines is the following:

You have gone to the doctor to get the results of a test. The test results show an alteration related to a very serious illness. Following the doctor’s instructions is successful in 20% of the cases. You feel moderately able to follow the doctor’s guidelines.

Each scenario was presented on a separate sheet of paper, and for each one the participants were asked to rate to what extent they would feel stressed in the conditions presented on the scenario. The participants responded on an 11-point Likert scale whose choices ranged from ‘hardly any stress’ to ‘extremely stressed’.

Design

According to the methodology of simulated scenarios, for each variable being studied, three different levels were considered, giving rise to a within-subject experimental design of 3x3x3 in which the scenarios were presented randomly. The use of a within-subject design ensures that all the participants experience all the conditions, which enables the control of possible individual differences.

Procedure

The procedure had two phases, where the participants rated the scenarios individually: a familiarisation one and an experimental one. Only the data obtained on the experimental phase were considered for analysis. In the familiarization phase, the participants rated 12 scenarios, two with the most extreme levels to avoid the ceiling and floor effect in the remaining scenarios, and ten more extracted at random from the total of 27 experimental ones. In this phase, the participants were able to ask questions to clarify the information. In the experimental phase, all 27 scenarios were presented randomly for each participant.

Data analysis

We conducted an analysis of variance (ANOVA) for repeated measures with the severity (mild, moderately serious and very serious illness), the efficacy of treatment (20%, 50% and 90%) and the perceived ability to follow the medical guidelines (unable, moderately able and very able) as the factors. For significant interactions, we used post-hoc pairwise tests with Bonferroni adjustment for multiple comparisons to test which of the changes in perceived stress due to one factor differs at each level of the other factor of the interaction.

With this analysis, and according to Anderson (1996), we can determine if the information is combined in an additive way (non-significant interactions meaning that the more variables at play the greater the effect in the resulting judgement) or a multiplicative way (significant interactions meaning that the effect of some of the variables is multiplied when combined with certain values in the other variables).

Results

The perceived stress level ranged between 1.87 and 9.55 points; therefore any floor or ceiling effects can be discarded.

The analysis of variance (see Table 1), indicated that the main effects of the three variables included in the scenarios were significant. Figure 1 shows that higher levels of perceived stress were related to a high level of the severity of the illness, F(1.53,231.19) = 580.32, p < .001), a low level of the efficacy of the treatment, F(1.64,247.92) = 266.57, p < .001 and of the patient’s perceived ability to follow the medical guidelines, F(1.5, 226.63) = 250.28, p < .001.

Table 1.

Analysis of variance for the repeated measures 3×3×3 design (severity of the illness, efficacy of the treatment and self-efficacy to follow the medical guidelines) in determining the perceived stress level.

Source	df¹	F
Severity of the illness	1.53^‡	580.32*
Severity of the illness within-group error	231.19^‡	(12.16)
Efficacy of the treatment	1.64^‡	266.57*
Efficacy of the treatment within-group error	247.92^‡	(4.39)
Self-efficacy	1.5^‡	250.28*
Self-efficacy within-group error	226.63^‡	(5.11)
Severity of the illness by efficacy of the treatment	3.54^‡	2.09
Severity of the illness by efficacy of the treatment within-group error	534.68^‡	(1.49)
Severity of the illness by self-efficacy	4	6.63*
Severity of the illness by self-efficacy within-group error	604	(1.24)
Efficacy of the treatment by self-efficacy	4	8.40*
Efficacy of the treatment by self-efficacy within-group error	604	(1.36)
Severity of the illness by efficacy of the treatment by self-efficacy	8	1.78
Severity of the illness by efficacy of the treatment by self-efficacy within-group error	1208	(1.12)

Note: Values enclosed in parentheses represent mean square errors.

p < .001

‡

Corrected F (Greenhouse-Geisser).

Figure 1.

Estimated levels of perceived stress as a function of severity of the illness, efficacy of the treatment and self-efficacy.

Regarding the additive and multiplicative effects between primary and secondary appraisals, the interaction between the severity of the illness and the self-efficacy was significant, F(4,604) = 6.63, p < .001, as was the interaction between the self-efficacy and the efficacy of the treatment, F(4,604) = 8.4, p < .001. In contrast, the interaction between the severity of the illness and the efficacy of the treatment was not significant, F(3.54,534.68) = 2.09, ns, nor the second-order interaction, F(8,1208) = 1.78, ns.

Additive relation between severity of illness and efficacy of the treatment

The non-interactive relationship between the severity of the illness and the efficacy of treatment to determine the perception of stress is shown in the left panel of Figure 2. The level of stress rose with an increase in the severity of the illness at all levels of efficacy of the treatment of the illness, and a lower the efficacy of the treatment also made the stress rise equally in all the levels of severity considered.

Figure 2.

Estimated levels of perceived stress as a function of severity of the illness and efficacy of the treatment (left), severity of the illness and self-efficacy (center) and self-efficacy and efficacy of the treatment (right).

Multiplicative relation between severity of illness and self-efficacy

The central panel of Figure 2 shows the perception of stress according to the severity of the illness and the self-efficacy. At all levels of severity of the illness, the lower the self-efficacy, the higher the levels of perceived stress, but the increase in the level of perceived stress observed between the moderate and the low self-efficacy levels was higher in the case of a very serious illness, F(1,151) = 134.95, p < .001, CI[0.90,1.27], than in a mild illness, F(1,151) = 63.15, p < .001, CI[0.51,0.84].

In contrast, the more serious the illness, the higher the levels of stress and this rise tended to be greater at the low self-efficacy level, F(1,151) = 788.71, p < .001, CI[3.85,4.43], compared to the high self-efficacy level, F(1,151) = 539.21, p < .001, CI[3.41,4.04].

Multiplicative relation between self-efficacy and efficacy of treatment

Regarding the significant interaction between self-efficacy and the efficacy of the treatment, the right panel of Figure 2 shows that higher efficacy of the treatment of the illness lowered the perceived stress, but that this effect rose as self-efficacy increased: the drop in perceived stress in illnesses with 90 percent effective treatment, compared to illnesses with 50 percent effective treatment, was lower when the self-efficacy was lower, F(1,151) = 64.48, p < .001, CI[0.55,0.91], than when the self-efficacy was moderate, F(1,151) = 160.81, p < .001, CI[0.93, 1.28], or high, F(1,151) = 196.96, p < .001, CI[1.09,1.45].

On the other hand, a higher self-efficacy lowered the perceived stress, but it did so more when the efficacy of the treatment of the illness was 90 percent, F(1,151) = 265.62, p < .001, CI[1.76,2.24], than when the efficacy of the treatment was 50 percent, F(1,151) = 168.17, p < .001, CI[1.24,1.69]. In contrast, a high self-efficacy lowered the perceived stress equally when the efficacy of the treatment of the illness was 50 percent or 20 percent, F(1,151) = 222.73, p < .001, CI[1.30,1.70].

Discussion

The purpose of this study was to ascertain how the severity of the illness, the likelihood of success of the treatment and the perceived ability to follow the medical guidelines combine in order to establish a judgement on the level of perceived stress regarding the information contained in a diagnosis. One of the first results to highlight is that all three factors proved to be relevant when studying the perceived stress caused by a health diagnosis. This information fully agrees with the main factors spotlighted by the majority of cancer patients when they are asked about the kind of information they want to receive (Fujimori and Uchitomi, 2009). In our study, the significance of the main effects indicates that the levels designed for each variable give rise to differing degrees of perceived stress (the stress is the highest when the severity of the illness is the highest, the efficacy of the treatment is low and the patient feels that he or she is unable to following the doctor’s instructions). This differing degree in the participants’ evaluation of stress validates the scenarios considered (Anderson, 1996).

The first hypothesis was confirmed by obtaining an additive relationship between the severity of the illness and the efficacy of the treatment. This kind of relationship between the primary appraisal and the outcome expectancy has been proven in similar contexts such as the study of preventive health behaviours. Specifically, the study by Frileux et al. (2004), analyses how different kinds of information combine to determine subjects’ intention to adopt preventive behaviour to fend off coronary disease. Their results show that the higher the threat and the more control the patient can exert, the greater this intention to behave better. In our study, despite the fact that the goal is to ascertain the degree of perceived stress, the same relationship between the threat or the severity of the illness and control or efficacy of the treatment is replicated.

In this sense, the three levels of treatment (90%, 50% and 20%) generate different degrees of perceived stress, and this effect on the efficacy over the perceived stress is sustained regardless of whether the illness is mild, moderate or serious. The perception of stress regarding the information in the diagnosis is higher when it is a serious illness and the efficacy of the treatment is low.

The second hypothesis was also confirmed, since a multiplicative relationship was observed between the severity of the illness and the perceived ability to follow the doctor’s instructions. Generally speaking, the perceived stress rises according to the severity of the illness such that if the illness is not important the perceived stress is also low regardless of whether the person feels unable or able to follow the doctor’s instructions. However, when the illness is serious, the levels of perceived stress rise, but in this case the effect of the perceived ability to follow the doctor’s instructions is relevant given that if it is low, this leads to a steeper rise in perceived stress than what is triggered by this low ability when the illness is not serious. Thus, unlike the efficacy of the treatment, which does not interact with the different levels of severity of the illness, the perceived ability to follow the doctor’s instructions takes on a differential role in triggering stress when combined with a given level of severity of the illness. In other words, the negative effect of perceived stress caused by the fact that the patient feels like he or she has few resources to follow the doctor’s instructions is particularly noticeable when the person is facing a serious illness. In a meta-analysis conducted by Franks and Roesch (2006) in which they examined the relationship between the cognitive appraisal of people with cancer and the kind of general coping strategy they use, it was noted that when the patients perceive their illness as a threat or loss they show problem-focused coping strategies and not avoidance or emotion-focused strategies, as their initial hypothesis posited. The authors suggest that this relationship between the perception of threat and problem-focused coping strategies might be influenced by aspects from the secondary appraisal, such as believing themselves to be able to follow the medical treatment, trusting in the support of family members or friends and seeing themselves as having the resources they needed in order to get the critical information on the illness and its treatment, among other considerations. Our results might support this proposal, since the primary appraisal of threat, represented in our case by the presence of a serious illness, moderates the levels of perceived stress when combined with a high perception of resources, which could encourage the use of more active coping strategies targeted at the illness, such as the ones noted in the aforementioned meta-analysis.

Regarding the third hypothesis, the expected multiplicative relationship between the perceived ability to follow the medical guidelines and the efficacy of the treatment was also confirmed: when the person believes that he or she has few resources to follow the medical guidelines, the fact that these guidelines have a 20 percent or 90 percent success rate does not make the difference in their perceived stress. At the other extreme, when the person believes himself or herself to be very able to follow the doctor’s instructions and therefore to benefit from the efficacy of the treatment, being able to pursue a course of treatment with a 90 percent success rate drastically lowers the level of stress compared to if the efficacy of the treatment is 50 percent or 20 percent. When these higher uncertainty values are present, the perceived stress does not change according to the individual’s perceived ability to follow the doctor’s instructions. Somehow, when the guarantee of success is low, this also limits the patient’s ability to control matters and therefore the importance of feeling more or less able to act drops.

In short, the patient’s beliefs of control, either regarding the efficacy of the treatment or their ability to follow it, interact with each other, confirming the evidence coming from laboratory studies regarding the effect that these variables have on the physiological parameters of stress (Sanz et al., 2006).

The contributions of this study should be understood in light of some limitations. Given that the sample was only made up of university students from a limited age bracket, these results would have to be replicated in other groups, given that this might influence the way they appraise and combine the information.

On the other hand, conducting future studies with a sample of participants with extreme scores on optimism might add information to the results found in this study, as they may display different patterns of combination. This variable may be the most important among individual dispositions, since there are many factors regarding health that are beyond the person’s direct control, and under these circumstances is where optimism has been proven to be most predictive of adjustment (Fernandez-Castro et al., 2009).

In conclusion the degree of perceived stress when receiving a diagnosis is the result of first the sum of the effects of the severity of the illness and the low efficacy of the treatment proposed, and secondly the multiplicative effect of feeling more or less able to follow the treatment. Regarding this latter effect, the lack of efficacy of the treatment raises the perception of stress proportionally to the different degrees of illness severity; on the contrary, having a low perceived ability to follow the treatment triggers high levels of stress, particularly when the diagnosis is serious.

From a clinical standpoint, having in mind the additive or multiplicative relationship between these components helps to know how to change the way bad news are communicated, in order to lower the perceived stress of a diagnosis. An additive relationship means that we can change any of the stressing information to lower the perceived stress of the patient; a multiplicative relationship means that we have to notice which one of the stressing information should be changed first, in order to decrease faster the perceived stress level. Thus, according to the additive results implying context factors related to the illness, when the diagnosis of a serious illness with few chances of recovery has to be communicated, it is important to follow the medical protocols published to communicate bad news based on giving the chance to receive personalized information on the benefits and risks entailed in following the medical guidelines, thus avoiding information that prompts uncertainty.

On the other hand, the multiplicative results implying self-efficacy contribute to this approach, gleaning the need to take into account information that does not depend on the context of the illness but that has to do with the patient’s personal resources. The patient’s level of confidence in his or her ability to follow the medical instructions should be encouraged, particularly when a serious illness is diagnosed. Needed resources should be offered, so that the patient can confidently cope with all the requirements of his or her illness, or ensure that he or she has them available. Clarifying questions or facilitating monitoring strategies when enhancing the health benefits of the recommended medical guidelines will lead to a drop in the perceived stress regarding the diagnosis. This result also reinforces the need of conducting new studies leading to know how other type of personal resources that influence perceived stress level, as the social support, combine its effects with context illness factors.

Footnotes

Funding

This study was supported by the grant [PSI2010,18397] from the Ministerio de Educación y Ciencia, Spain.

References

Anderson

(1996) A Functional Theory of Cognition. New Jersey: Lawrence Erlbaum Associates.

Baile

Buckman

Lenzi

Glober

Beale

Kudelka

(2000) SPIKES-A six-step protocol for delivering bad news: Application to the patient with cancer. Oncologist 5(4): 302–311.

Bandura

(1997) Self-efficacy: The Exercise of Control. New York: Freeman and Company.

Bargiel-Matusiewicz

Trzcieniecka-Green

Krupa

Krzystanek

(2005) Reaction phases following HIV positive diagnosis. Archives of Psychiatry and Psychotherapy 7(1): 63–69.

Beadle

Yates

Najman

Clavarino

Thomson

Williams

. (2004) Beliefs and practices of patients with advanced cancer: Implications for communication. British journal of cancer 91(2): 254–257.

Buckman

(1984) Breaking bad news: Why is it still so difficult? British Medical Journal (Clinical research edn) 288(6430): 1597–1599.

Clarke

Stone

(2007) Adjusting to illness and other major life events - How can GPs help? Australian Family Physician 36(4): 206–210.

de Haes

Bensing

(2009) Endpoints in medical communication research, proposing a framework of functions and outcomes. Patient Education and Counseling 74(3): 287–294.

Fernandez-Castro

Rovira

Doval

Edo

(2009) Optimism and perceived competence: The same or different constructs? Personality and Individual Differences 46(7): 735–740.

10.

Franks

Roesch

(2006) Appraisals and coping in people living with cancer: A meta-analysis. Psycho-Oncology 15(12): 1027–1037.

11.

Frileux

Munoz Sastre

Mullet

Sorum

(2004) The impact of the preventive medical message on intention to change behavior. Patient Education and Counseling 52(1): 79–88.

12.

Fujimori

Uchitomi

(2009) Preferences of cancer patients regarding communication of bad news: A systematic literature review. Japanese Journal of Clinical Oncology 39(4): 201–216.

13.

Lazarus

(1991) Emotion and Adaptation. New York: Oxford University Press.

14.

Lazarus

Folkman

(1984) Stress, Appraisal and Coping. New York: Springer Publishing Co.

15.

Lee

Back

Block

Stewart

(2002) Enhancing physician-patient communication. Hematology American Society of Hematology Education Program: 464–83.

16.

Lusar

Faixa

Fernández-Castro

(2010) Juicios de riesgo en el deporte: Una aproximación experimental en excursionistas. Revista de Psicología del Deporte 19(2): 203–217.

17.

Mager

Andrykowski

(2002) Communication in the cancer ‘bad news’ consultation: Patient perceptions and psychological adjustment. Psycho-Oncology 11(1): 35–46.

18.

Munoz Sastre

Mullet

Sorum

(2000) Self-assessment of inebriation from external indices. Addictive Behaviors 25(5): 663–681.

19.

Ptacek

Eberhardt

(1996) Breaking bad news - a review of the literature. Jama-Journal of the American Medical Association 276(6): 496–502.

20.

Randall

Wearn

(2005) Receiving bad news: Patients with haematological cancer reflect upon their experience. Palliative Medicine 19(8): 594–601.

21.

Rovira

Edo

Fernandez-Castro

(2010) How does cognitive appraisal lead to perceived stress in academic examinations? Studia Psychologica 52(3): 179–192.

22.

Sanz

Villamarin

Alvarez

(2006) Effects of specific and non-specific perceived control on blood pressure response in a stressful mental task. Biological Psychology 71(1): 20–28.

23.

Sharf

Stelljes

Gordon

(2005) ‘A little bitty spot and I’m a big man’: Patients’ perspectives on refusing diagnosis or treatment for lung cancer. Psycho-Oncology 14(8): 636–646.