Knowledge and attitude towards advance directives for patients with terminal illnesses among doctors working in a tertiary care hospital

Study design and setting

This cross-sectional survey was done in a tertiary care hospital with a bed strength of 2700. This teaching hospital admits 100 medical undergraduates and about 500 postgraduates (broad and super specialty) each year for training. Apart from the trainees, the hospital employs around 1300 medical doctors.

Sample

All interns, postgraduate trainees, and faculty from both clinical and non-clinical departments were eligible for the study. Every fifth person on the list of eligible participants was contacted for enrolment, either in person or through email. At least 3 reminders were sent to each individual before they were considered a non-responder, after which the next name on the list was contacted.

Ethical considerations

The Institutional Review Board and Ethics Committee approved the study protocol.

Assessment

Socio-demographic, academic, and professional details were collected using a specially designed proforma. The participants’ religiosity was explored using questions adapted from the Centrality of Religiosity Scale (CRS). This 5-item scale has a good internal consistency (D: 0.85).²⁵ The knowledge and attitude questionnaire was used from a previous study from Saudi Arabia.²² This scale has good psychometric properties (knowledge scale reliability: α=0.88 and attitude scale reliability: α=0.93). The knowledge section of the questionnaire has 12 questions related to the definition and types of AD, living will, durable power of attorney, onset of AD’s validity, itemising of several clinical practices into the AD, ideal timing of AD discussion, nomination of a person as healthcare proxy, and the incorporation of that person in the discussion of AD. The answers are scored 0 or 1, indicating absence or presence of knowledge, respectively. The attitude section of the questionnaire contains 27 questions based on 4 sub-themes: planning of AD, comfort and confidence in discussing AD, application of AD, and challenges of AD. The responses of ‘No’ and ‘Yes’ were scored 0 and 1, respectively. The ‘I don’t know’ option was available for some questions. One question was modified, wherein ‘Indian legal system’ replaced ‘Islamic regulations’ from the original questionnaire. A modified Bloom’s cut-off of 75% was used to dichotomise participants’ knowledge into good or poor, and participants’ attitudes as positive or negative.²⁶

Statistical methods

The sample size was calculated based on a 68% prevalence of positive knowledge from a previous study.²² Using the statistical formulae for hypothesis testing for two means, knowledge score, and two proportion hypothesis testing (large proportion— equal allocation) and assuming a 10% dropout rate, the sample size was calculated as 400—this included 200 doctors currently undergoing training (interns and postgraduates), and 200 doctors who had completed training in their specialty. Study participants were selected by systematic probability sampling. Mean, standard deviation, and range were used to describe continuous variables, while frequency distributions were obtained for categorical variables. Parametric tests were used in the analysis as the data were normally distributed. The chi-square test and Student’s t-test were used to test for associations of categorical and continuous variables, respectively. Pearson’s correlation coefficient was used to test for the association between two continuous variables. Multivariate linear and logistic regression analysis was done using factors found significant in the bivariate analysis. SPSS for Windows (version 16.0.1) was used for analysis of data.

Knowledge of AD

Less than half (43.5%) of the study participants had good knowledge regarding AD. More post-training participants had a good total knowledge score (49.4%) as compared to those in training (38.8%). Knowledge about the definition of AD and overall knowledge was significantly better among post-training participants. Knowledge regarding details of withholding nutrition and hydration in AD and the ideal time for discussing AD was found to be poor among both groups (Table 2).

Training in AD (p<0.001) and experience in managing terminally ill patients (p=0.002) were found to be associated with higher knowledge scores, and remained significant when adjusted for confounders in multivariate logistic regression (training status, family history of terminal illness, training in AD, experience in working among terminally ill were entered into the model; Table 4).

Attitude towards AD

A positive attitude towards AD was found in 33.2% of the entire sample; among the post-specialty-trained doctors and in-training group, rates of positive attitudes were 32% and 34.2%, respectively. Positive attitudes were lowest in the comfort and confidence subscale of the questionnaire (in-training 28.5%, post-training 28.3%). Attitudes were predominantly negative for questions pertaining to ease of discussion about end-of-life care and AD, confidence in breaking ‘bad news’, and the cultural and legal aspects of AD. Significant differences between the groups were noted in individual attitude questions related to the impact on patients’ hope, AD precipitating a confrontational relationship with the patient, AD reducing the incidence of futile and unnecessary end-of-life care, and the option of recommending AD in the institute, with post-training participants having more positive attitudes (Table 3).

High religiosity (p=0.03), training in AD (p=0.02), working in surgical specialty (p=0.001), working in a superspecialty department (p=0.02) and adequate knowledge (p<0.001) were associated with attitudes towards AD. In multivariate logistic regression analysis of the predictors of positive attitude towards AD, area of work and adequate knowledge remained significant when adjusted for other confounders (level of religiosity, area of work, training in AD, and knowledge on AD were entered into the model; Table 4).

Among doctors working in superspecialty departments, number of years of experience after training positively correlated with attitude subscales of comfort and confidence in discussing AD (r=0.33, p=0.04), application of AD (r=0.34, p=0.03), challenges of AD (r=0.39, p=0.01) and total attitude scores (r=0.37, p=0.02).

Knowledge about AD

Less than half of the participants in this study had adequate knowledge regarding AD, similar to the results obtained by some researchers.^{22,24,31–33} Similar to some previous reports, in this study, knowledge was inadequate in the domains related to withholding of nutrition and hydration, and the ideal time to discuss AD.²² As expected, prior training in AD and working with terminally ill patients were associated with adequate knowledge. These findings have also been reported previously.^22,33

Attitude towards AD

Two of every 3 medical doctors in this study reported a negative attitude towards AD. Similar prevalence rates have been reported in earlier studies.²² This high rate may be contributed to by the paternalistic attitudes prevalent among medical professionals in this country. It may also be a consequence of inadequate exposure to the concept of AD, which are a recent addition to medical ethics in our country. What is heartening is the positive correlation of positive attitudes with knowledge. This suggests that attitudes to AD can be improved with training, as also suggested by some previous reports.^21,22,33

Some religious scholars believe that the concept of AD is not congruent with their religious teachings.³⁴ The bulk of the available literature on AD is from countries where the Abrahamic religions predominate. We attempted to study the interaction between religious affiliation and attitudes and knowledge on AD. Participants who stated that they were highly religious were found to be more likely to have negative attitudes towards AD. This is similar to a previous report that suggested that physicians who were very religious were less likely to believe that life-sustaining treatment should be withdrawn, or to endorse the prescription of needed pain medication if it hastens death, or agree with euthanasia.³⁵ Other researchers have not found a link between religion and AD.^36,37 Al-Jandali et al. have discussed ways to address the religious implications of end-of-life decisions.³⁶

Our findings highlight the deficiencies in the medical curriculum in imparting ethical concepts, especially when dealing with the terminally ill. Education regarding the indications for ACP and implementation of AD would improve the quality-of-care for the terminally ill. These aspects could be included as modules in the undergraduate and postgraduate curriculum, or as continuing medical education programs.¹⁹ The National Medical Commission flagship program for medical faculty— AETCOM—includes the concept of AD among the core competencies; however, it is not elaborated in the modules. A greater emphasis on promoting attitudes and skills would be one of the first steps toward improving medical teachers’ competency.³⁸ Training should initially focus on medical professionals in specialities where ACP and AD are applicable in routine clinical practice, such as geriatrics, oncology, palliative care, etc. Future measures to incorporate ACP into the Indian healthcare delivery system will require a cohesive plan among all medical and surgical specialties. Care planning implementation should also extend beyond the healthcare setting into the community, including social and community health workers, legal professionals, and faith-based and community leaders.⁵ The general public needs to be involved through the organization of awareness programs. Non-profit organizations, nongovernmental organizations, and religious bodies can be asked to help disseminate the knowledge and awareness about the concept of ACP in the community. This bottom-up approach will ensure better translation from a theoretical and legal framework to clinical and practical output.¹⁹ AD should undergo standardization and validation and then be stored in a manner allowing the information to be accessible throughout India, confidentially. The cultural and religious perspectives of the providers and recipients of ACP and AD must be explored. In the near future, care planning must be reconceptualised as an ongoing process through the course of life that includes immediate and advanced decisions.^5,19

Strengths and limitations

This survey is one of the first explorations into a relatively new but important ethical concept in the country. An adequate sample size ensures the study is well powered to draw meaningful conclusions about the population studied. The use of a previously validated structured instrument to assess knowledge and attitudes towards AD is another strength.

Several limitations merit mention. Our study was done in a multilingual minority institution; hence, the study population may not be representative of the medical fraternity in India, with respect to demographics or culture. However, the professional characteristics are representative of the faculty of any teaching hospital in India. The inclusion of physicians and trainees from non-clinical specialities may have skewed the results, as specialists from non-clinical departments do not routinely interact with patients with terminal illnesses. The cross-sectional nature of the study precludes us from understanding the directionality of association or changes in knowledge and attitude over time. A qualitative approach with research interviews on this topic may have obtained greater detail about knowledge and attitudes regarding AD in this group. Future research should focus on assessing changes in knowledge and attitude regarding AD at different times in a medical professional’s career, as well as among different groups of professionals, such as nurses and paramedical workers. Continuing medical education (CME) sessions with pre- and post-test assessments would allow for the evaluation of the effectiveness of such programs. The results of such research would also help formulate legislation and standard operating procedures regarding AD.

Conclusions

Medical ADs are a novel concept in India, and discussions about them are still in the early stages. It is essential to explore the level of knowledge and attitudes regarding AD among medical professionals, patients, and caregivers, prior to conceptualizing legal and medical protocols for the same. Among medical professionals, competence in discussing and initiating AD should be encouraged from the undergraduate training period onward. Medical specialities such as palliative care, oncology, psychiatry, and geriatrics should ensure that postgraduate trainees are able to adequately assess competency and decisional capacity, and are knowledgeable about the ethical and legal aspects of AD. Strong ethical and legal departments are essential in healthcare institutions to ensure transparency in the planning and implementation of advance directives, which are an important end-of-life concern for patients and caregivers.