Subjective expectations regarding longevity and future health: a cross‐sectional survey among patients with Crohn's disease

The aim was to explore the subjective health expectations (sHE) of patients with Crohn's disease (CD) for both the near future and the elderly.


Introduction
Crohn's disease (CD) is a chronic inflammatory bowel disease that may affect any part of the digestive track causing abdominal pain and diarrhoea, and may lead to functional disability and to life-threatening complications. Living with CD may cause a substantial burden on patients due to alterations in their current health, but not less importantly due to concerns and uncertainties they may have regarding their future health [1]. Patients' preferences and expectations regarding treatments and healthcare have come into focus in the past years [2][3][4]. However, still little is known about how conscious CD patients are about the chronicity and course of their disease and its impact on their future health, yet these can play a key role in their current lifestyle and health-related decisions [5,6].
Brouwer and colleagues highlighted that individuals' subjective health expectations (sHE) may differ from objective measures and inaccurate subjective expectations regarding length and quality of life may have important implications for health behaviour [7][8][9]. Those who underestimate their future health may place less importance on seeking medical care and be less compliant. In contrast, overestimation of future health can influence negatively the perception of actual treatment effects and may lead to dissatisfaction with care. These issues are especially relevant and can be deterministic for successful care in chronic diseases, such as CD, that require long-term care and compliance.
The aim of our study was therefore to explore CD patients' sHE regarding their health status 1 year ahead and for ages 60, 70, 80 and 90, as well as their beliefs on longevity. Secondarily, we aimed to analyse determinants of sHE and compare CD patients' results with previous studies among the general public and other chronic diseases.

Method Study design
A cross-sectional survey was conducted at three academic gastroenterology departments and an inflammatory bowel diseases centre in Hungary. Details of the study have been described elsewhere [10,11]. In brief, consecutive outpatients aged 18 years and over, with a definitive diagnosis of CD, were invited to participate in the study. Ethical approval for conducting the study was granted by the National Scientific and Ethical Committee (reference no. 49548-4/2016/EKU) and the study was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. Informed consent was signed by all patients. A paper-based questionnaire survey was performed in which patients were asked about their socio-demographic characteristics (self-reported sex, date of birth, educational level, residence, employment status), CDrelated symptoms and health status. Gastroenterologists provided data on patients' medical history, clinical characteristics, disease severity and treatments.

Measurement tools
Crohn's Disease Activity Index (CDAI) CDAI is a measure of disease activity in CD [12]. It includes signs and symptoms based on the past 7 days, and consists of eight components with a weight factor attached to each. The total score is calculated by the physician as a sum of items multiplied by weighting factors. Cut-offs for remission or low disease activity, mildly active, moderately active and severely active disease are < 150, 150-219, 220-449 and 450-600, respectively [13].

Perianal Disease Activity Index (PDAI)
The PDAI measures the severity of perianal fistulising CD, covering five dimensions: fistula discharge, pain and restriction of activities, sexual activity restriction, type of perianal disease, and degree of induration [14]. Each dimension is scored on a scale between 0 (no symptoms) and 4 (severe symptoms); the total PDAI score is calculated as the sum of the items (range between 0 and 20). PDAI ≤ 4 indicates inactive disease while PDAI > 4 is considered as active disease [15].

Visual analogue scale (VAS)
The Patients' Global Assessment VAS (PGA VAS) of 0-100 with end-points defined as 'not severe at all' (0) and 'very severe' (100) was applied to assess patients' self-perceived severity of CD in general. CD-related current pain and the worst experienced CD-related pain in the past 3 months were measured separately with a 0-100 VAS (pain VAS and worst pain VAS, respectively), with end-points of 'no pain at all' (score 0) and 'pain as bad as it could be' (score 100).

EQ-5D-3L questionnaire
The EQ-5D-3L is a generic health status measure consisting of two parts [16,17]. The first part is a descriptive system covering five dimensions of health: mobility, selfcare, usual activities, pain/discomfort and anxiety/depression. Respondents are asked to indicate on a threelevel response scale (1no problem; 2some/moderate problem; 3unable/extreme problem) for each dimension that best describes their current health status. Preference weight (utility) originating from population-based studies can be attached to each health state description. Due to lack of country-specific tariffs in Hungary, the value set of the UK by Dolan was used (score range from À0.549 to 1.0) [18]. The second part is a vertical 20 cm visual analogue scale (EQ VAS) with end-points of 0 indicating the worst imaginable health status and 100 indicating the best imaginable health status.

Assessment of subjective expectations on future health and longevity
Patients were asked to indicate on the descriptive system of the EQ-5D-3L questionnaire the health status they expect to have a year later. To assess sHE for future ages, we followed the same methodology that has been successfully used in previous studies [7,8,[19][20][21][22] (Table S1). Patients were asked to indicate the level of health problems they think they will have at ages 60, 70, 80 and 90 in the five dimensions of the EQ-5D-3L. The EQ-5D-3L index scores for the subjectively expected health states were calculated. Patients were also asked to indicate the age to which they expected to live (subjective life expectancy, sLE).

Statistical analysis
The difference between patients' sLE and age-and sexmatched statistical life expectancy was computed based on data retrieved from the Hungarian Central Statistical Office (latest available data: year 2016) [23]. The differences between expected (at year 1) and current EQ-5D-3L index scores and between patients' subjective and statistical life expectancy were tested by Wilcoxon signed-rank test. sHE across subgroups of patients were compared by using the Mann-Whitney U test or Kruskal-Wallis test, as appropriate. The relationship between the continuous variables was analysed by Spearman's correlations. A correlation coefficient of 0-0.19 was defined as very weak, 0.20-0.39 as weak, 0.40-0.59 as moderate, 0.60-0.79 as strong and 0.80-1 as a very strong correlation. All the applied statistics were twosided with a significance level of P < 0.05. Statistical analyses were performed with IBM SPSS version 25.0 (SPSS Inc., Chicago, Illinois, USA).

Results
Two hundred and six patients (54.9% men) with a mean age of 34.7 (SD 10.5, range 18-70, median 34) years participated in the survey. There were 31 patients in the age group 45-54, but only seven patients in age group 55-64 and one aged over 65; therefore we joined these three age groups for the analyses. Main clinical and socio-demographic characteristics of the sample are presented in Table 1. The average disease duration was 10.5 (SD 6.3; N = 205) years, the mean CDAI score was 110.5 (SD 77.0; N = 206) and the PDAI score was 3.7 (SD 2.3; N = 83). The PGA, current and worst pain in the past 3 months VAS scores were on average 50.1 (SD 23.9; N = 204), 24.7 (SD 23.9; N = 206) and 42.0 (SD 31.4; N = 206), respectively. The majority (75.7%) of the patients had low disease activity or were in remission (CDAI < 150) and 66.0% were treated with a biologic drug at the time of the assessment: infliximab 46.6%, adalimumab 17.5%, vedolizumab 1.9%. Current general health status as measured by the EQ-5D-3L index was on average 0.8 (SD 0.17; N = 203) and 72.7 (SD 19.7) on the EQ VAS.
Patients' subjective health expectations for 1 year ahead and for ages 60, 70, 80 and 90 Patients expected a modest improvement (mean 0.05, SD 0.15, P < 0.05) in their health within a year on the EQ-5D-3L index (Table 1). Women expected somewhat greater improvement than men but the provisioned change was statistically significant in both subgroups. The highest difference between the current and expected EQ-5D-3L index score was observed in age group 18-24 years.
Patients expected a significant worsening of their health status with age achieving a mean score of À0.12 (SD 0.52) at age 90 ( Table 1). The expected health status for older ages did not differ significantly by sex, educational level, clinical subtype (fistula), present treatment type or by having had or not CD-related surgery in the past. In contrast, the difference was statistically significant across age groups for all the four future ages. Those who had had extraintestinal manifestation expected significantly worse health status for ages 60 and 70. We did not find a significant difference between subgroups who expected or not to reach the specific future ages in question.

Patients' subjective expectations for longevity
The main findings are presented in Table 2. Patients with CD expected to live up to 76.8 (SD 3.5; range 40-100) years of age on average. Women and patients in age group 35-44 expected on average 3.0 (SD 12.0) and 4.2 (SD 10.6) years shorter life than their sex-and age-matched statistical life expectancy, respectively. Significant but weak negative correlations were found between sLE and current general health and pain (EQ-5D-3L, PGA VAS, pain VAS), but not with the diseasespecific CDAI and PDAI scores. Correlations between sLE and sHE for 1-year ahead and for future ages were weak and moderate, respectively (Table 3).

Correlations between subjective health expectations and patient characteristics
Results are presented in Table 3. With regard to sHE in 1 year's time, we found that patients who were younger, had shorter disease duration and better general health status (EQ-5D-3L) expected to have a significantly better EQ-5D-3L index score. Stronger CDrelated pain (currently and in the past 3 months) negatively influenced the expected EQ-5D-3L score for The relationship was not significant with either the CDAI or the PDAI scores. Our findings with respect to sHE at ages 60, 70, 80 and 90 were very similar, with a few exceptions (Table 3).

Discussion and conclusions
In this empirical study we surveyed the sHE of adult CD patients attending for outpatient care at gastroenterology departments. Patients' current general health status was somewhat worse than that of the general population in Hungary (mean EQ-5D-3L index score of 0.80 at a mean age of 34.7 years vs 0.86 in age group 35-44 years) [24]. Patients expected on average a statistically significant improvement in their general health status (0.05-point increase on the EQ-5D-3L) within a year. To the best of our knowledge, a minimum clinically important difference for the EQ-5D-3L index score has not been established in CD, yet [25]. Nonetheless, we think this result indicates that patients expect rather a stability (non-worsening) and maybe some improvement of their disease in the near future. In contrast, CD patients provisioned a sharp deterioration of health in their older age (Table 1). CD patients expected to have worse health status than the general population at these ages (Fig. 1). The gap between short-term and long-term sHE suggests that CD patients attending gastroenterology care are rather optimistic regarding their health in the immediate future but they have serious concerns about their chances to have well-controlled disease over a lifetime. Qualitative studies are suggested to explore the underlying causes, including medical, social (access, affordability) and patient-related reasons [26,27]. Clinicians should make efforts to explore patients' sHE and help them to put their sHE on a realistic perspective based on the available evidence. On the one hand, overestimation of short-term treatment effects may lead to disappointments regarding the results achieved and may induce feelings of inadequacy and lead to worsening of compliance. On the other hand, a pessimistic vision about the distant future is probably an unfounded extra burden and avoidable harm on most CD patients.
Our analyses revealed that age, disease duration, selfreported current health (EQ-5D-3L) and pain measures correlated significantly with 1-year sHE but disease activity measures (CDAI, PDAI) did not (Table 3). Current health correlated significantly also with sHE for ages 60, 70 and 80. Hence applying these patient-reported outcomes alongside the standard medical decision making tools (CDAI and PDAI) is strongly suggested as these carry valuable information not only about patients' current self-perceived status but also about their beliefs on health perspectives. Most of the patients did not expect significant shortening of their life due to CD although an increased standardized mortality ratio may occur, especially in specific subgroups with CD [28]. sLE showed significant correlations with sHE. Therefore, it is strongly suggested to explore these two in parallel.
Comparisons with some previous studies deserve mentioning. sHE for future ages were explored in two online surveys among the general population in Hungary [19,29]. Both confirmed that people in general expect health deterioration with age and the expected health status is worse than the actual health status of the general public from age 70 and above. The same pattern was observed among the general population in the Netherlands [7]. Two separate studies in Hungary involving patients with rheumatoid arthritis and psoriasis reported very similar findings to ours in CD [20,21] (Fig. 1). It seems that, despite the incredible development that health professionals have witnessed in the past two decades in the treatment of inflammatory diseases in rheumatology, dermatology and gastroenterology, many patients still have little confidence in having good health in old age. Another similarity is that rheumatoid arthritis patients expected a robust improvement in health in the short term at the initiation of their first biologic treatment. Nonetheless, it was proved to be an overestimation based on the measurements at the follow-up visit [20]. Hence we think that more attention should be paid to informing patients about the awaited size, timing and character of the treatment effects, as well as about long-term effects of adequate care. To the best of our knowledge, similar studies have not been performed for inflammatory bowel disease patients.
Some limitations of our study deserve mentioning and we would like to suggest a few directions for further research. Patients in remission and with mildly active disease were overrepresented in the sample while only few patients aged 55 and over participated in the study. Given the cross-sectional design we could not match sHE to measured data. Therefore, we suggest testing the results of this explorative research in larger prospective studies involving representative patient samples and analysing the determinants of sHE. It is recommended to record response rate (which we have omitted) in order to assess potential non-response bias. We suggest considering the use of the EQ-5D-5L version (with five-level response scale) to explore sHE in future studies [30]. The EQ-5D-5L has been proved to have better measurement properties as a health status measure than the EQ-5D-3L [10], and we expect an increase in its use in CD registries and clinical studies. We used the tariffs of the UK to calculate actual and expected EQ-5D-3L scores due to lack of local tariffs at the time of the analyses. These tariffs may differ significantly across countries [31], and indeed, according to our preliminary analyses using the recently published country-specific tariffs for Hungary [32] we found somewhat different scores (slightly lower average score for current status and higher for 1 year ahead, and it was decreasing for future ages but at varying degrees; data not shown). For comparability reasons, we stayed with the UK value set in this current study. It would be useful also to investigate the potential disagreements between patients' and clinicians' views. Patients' sHE regarding treatment effects may change with the number of therapy failures and thus influence the evaluation of subsequent treatments. Influencing factors of sHE have only been partly identified in our study; further research (on health literacy, personality traits, cultural characteristics, availability of health and social care support) is recommended.
In conclusion, our study suggests that CD patients attending gastroenterology care expect health improvement in the near future but a severe health deterioration in later life. Patients' age, current general health status and pain seem to have a significant impact on sHE, but clinical history and current treatment might have substantial influence as well. We believe that our research has opened valuable new doors in the care of CD patients and hope it will inspire new sHE studies.  Marianna Rutka Eszter Sch€ afer, Tam as Szamosi, Ferenc Zsigmond and Zsuzsanna V egh for their contribution to data collection. Special thanks to Tam as Cserni for converting the paper-based data into an electronic dataset and assisting with the data analyses.

Conflicts of interest
M arta P entek is member of the EuroQol Group, a notfor-profit organization that develops and distributes instruments that assess and value health. The other authors report no conflict of interest.

Funding
This research got no funding.

Ethics approval
Ethical approval for conducting the study was granted by the National Scientific and Ethical Committee of Hungary (reference no. 49548-4/2016/EKU). The study has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Patient consent
An informed consent form was signed by all patients before the survey.
Permission to reproduce material from other sources Not applicable.

Clinical trial registration
Not applicable.

Data availability statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.

Supporting Information
Additional Supporting Information may be found in the online version of this article: Table S1. Questions used to explore subjective health expectations for future ages.