Participants
We conducted a survey of a convenience selection of 761 mental healthcare professionals assisting children with intellectual development disorder (IDD), autism spectrum disorder (ASD), and attention deficit hyperactivity disorder (ADHD), in order to explore their personal and professional characteristics in relation to their patterns of communication with the children’s parents. We invited psychiatrists, psychologists, nurses, social workers, and residents in psychiatry and related-healthcare areas to participate. The healthcare professionals worked in mental, neurological and children’s hospitals in Mexico City (Psychiatric Hospital Fray Bernardino, National Institute of Neurology and Neurosurgery, National Institute of Psychiatry Ramón de la Fuente Muñiz, Children's Psychiatric Hospital Dr. Juan N. Navarro). In addition, we included residents from the Medical School of the National Autonomous University of Mexico, paediatricians from the Children’s Hospital (central state of Morelos), and psychiatrists from the psychiatric healthcare services of Mexico City and the Jalisco Institute of Mental Health. At each institution, an in-person interviewer invited mental healthcare professionals to participate. The interviewer explained the objective of the study and the fact that participation was voluntary and the information provided would be confidential. Subsequently, informed consent was obtained from those who agreed to participate, and they were given the questionnaire, which was collected in the following two hours or the next day. The study was carried out from June 2018 to January 2019.
Questionnaire
"Patrones de comunicación de profesionales de la salud con padres de sujetos con: Trastorno del desarrollo intelectual (TDI), Trastorno del espectro autista (TEA), y Trastorno del déficit de atención-hiperactividad (TDAH)" ("Communication patterns of health professionals with parents of subjects with: Intellectual development disorder (IDD), Autism spectrum disorder (ASD), and Attention-hyperactivity disorder (ADHD)"), an instrument in Spanish, was used [27]. The instrument contains 64 items and is composed of two sections: a) professional and personal reflections and b) case studies or clinical vignettes. The personal reflections section corresponds to questions that explore situations the professionals encounter in both medical care and their daily lives; the responses were used to determine attributes such as paternalism, the value that professionals place on truth, their attitudes towards death, and their communication patterns. The clinical vignettes section presents case studies and includes questions regarding diagnosis, prognosis and treatment, which were used to construct indicators of the mental health professionals’ knowledge about IDD, ASD, and ADHD. In addition, the questionnaire included variables related to educational level and specialty, as well as religion and bioethics training. The questions were answered using a Likert-type response, which allowed an understanding of the mental healthcare professionals’ level of agreement with the proposed statements (strongly disagree, disagree, agree, and strongly agree). The questionnaire is an adapted version of an original instrument previously used in other studies of Mexican populations [28, 29] and has shown adequate internal consistency (0.76) through the Kuder-Richardson test [30]. When the questionnaire was developed, an expert panel evaluated the relevance and clarity of the selected items after three rounds of review.
Primary outcome
Paternalism was defined as an attitude and behaviour in which mental healthcare professionals impose their outlooks and decisions on their patients, limiting patient autonomy with the belief that they do so for the benefit of their patients or themselves. Paternalism (dependent variable) was constructed based on the following questions:
(a) The reaction that I want to inspire in my patients diagnosed with a chronic disease is 1- Confidence and calmness, 2- A combative spirit, 3- Active participation, 4- I do not intervene in the moods of my patients.
(b) The best hope we can give to a parent with a child diagnosed with IDD/ASD is to make him/her feel that life can continue as normally as possible.
(c) Emotional distress does very little; therefore, I try to assist the children’s parents as much as possible by avoiding feelings such as sadness, grief or anguish.
(d) Enthusiasm should be shared to encourage parents, even if it means telling a lie.
(e) We create the reality of others. For example, if a parent with a son diagnosed with an incurable disease sees me being calm, the parent will think, "If the physician is calm, the situation might not be so bad".
(f) When I see someone looking crestfallen, my first reaction is to try to distract that person to encourage him/her, even if it requires changing the subject.
(g) Talking about painful topics only makes the pain worse.
(h) When I have a problem, I try to conceal it from my loved ones.
(i) I was always taught to avoid causing someone distress.
First, the answers to each of the questions were addressed; those indicating that the professionals were in favour of low paternalism/autonomism were considered correct. Subsequently, a score was assigned; for example, if the professional fully agreed with one of the questions that indicated low paternalism, 4 points were assigned in accordance with the Likert scale score (strongly agree, 4 points; agree, 3 points; disagree, 2 points; strongly disagree, 1 point.) We determined the arithmetic sum of the scores for each item, and based on their distribution (tertiles), determined the following categories of paternalism: high paternalism/overprotection (T1, reference category), moderate paternalism (T2) and low paternalism/autonomism (T3).
Independent variables
The possible predictors of paternalism analysed were: a) communication pattern, which was defined as the behavior reported by mental health professional in relation to its communication style with parents when discussing the diagnosis, prognosis and/or treatment of patients with IDD, ASD and ADHD. To construct this indicator, 11 questions from the instrument were selected based on input from experts (items 1, 2, 6, 8, 9, 10, 11, 18, 33, 34, 40); b) Value assigned to the truth, which refers to the value the healthcare professional indicated placing on conveying the truth in his/her communication with parents; in other words, the correspondence between what the healthcare professional knows about the situation and what the healthcare professional tells the parents (items 18, 19, 28, 31); c) Attitude towards death, which refers to healthcare professional’s willingness to adapt, react and act in situations related to death (items 28, 29, 30, 41, 43, 44, 47, 49); d) Family member with IDD or ASD, which asks whether any member of the professional's family has been diagnosed with IDD or ASD (item 13); e) Bioethics courses, which refers to courses related to medical ethics that the healthcare professional has taken throughout his/her professional training (item 16); and f) Religion, which refers to whether the professional describes him/herself as a believer or nonbeliever regarding religion (item 53). To construct the communication pattern, value assigned to the truth and attitude towards death indicators, we followed the same methodology that was used for the paternalism indicator.
This study also included indicators related to knowledge about IDD, ASD, and ADHD as predictors of the paternalism. To construct these indicators, three clinical vignettes were presented in the last section of the questionnaire. The clinical vignettes were presented as cases or scenarios featuring people of a specific age with IDD, ASD or ADHD and were accompanied by different questions about diagnosis, prognosis and treatment [27]. Answers were considered correct if they were among those selected by the group of experts in paediatric psychiatry. For IDD and ADHD, 3 or 4 correct answers indicated a positive attitude and a high degree of knowledge, 2 correct answers indicated intermediate knowledge, and 1 or no correct answer indicated a low level of knowledge. For ASD, 3 correct answers indicated a positive attitude and a high level of knowledge, 2 correct answers indicated an intermediate level of knowledge, and 1 or 0 correct answer indicated a low level of knowledge. The inclusion of knowledge variables is important because they relate to the formulation of an accurate diagnosis of the mental illness being studied. Having an accurate diagnosis increases the health professionals’ confidence in communicating and discussing the disorder with either the patient or his or her parents [31-34].
Statistical analysis
A descriptive analysis of the study population was carried out. For comparisons, we used chi-square tests. To evaluate the association between the communication attributes and low paternalism/autonomism, a logistic ordinal multivariate model was constructed. Odds ratios (ORs) and 95% confidence intervals (CIs) were obtained. The following variables were considered possible predictors of low paternalism/autonomism: a) age (tertiles: 43-76 years as the reference category, 30-42 years, 19-29 years); b) gender (male, female); c) specialty (no, yes), d) value assigned to the truth (low, moderate, high); e) communication pattern (withholding, partial communication, open communication-understood as the communication style for which the professional obtained the highest scores, regarding to provide the most information to parents when discussing the diagnosis, prognosis and/or treatment of patients with IDD, ASD and ADHD); f) religion (nonbeliever, believer); g) attitude towards death (low acceptance, moderate acceptance, high acceptance); h) family member diagnosed with some type of IDD/ASD (yes, no); i) bioethics courses (none, ≥1), and j) knowledge about IDD, ASD and ADHD (low knowledge, intermediate knowledge, and positive attitude and high knowledge). To assess the joint effects of age and communication patterns or specialty on the likelihood of presenting low paternalism/autonomism, we created the following interaction terms: a) age (tertiles) and communication patterns (withholding, partial communication and open communication); b) age (tertiles) and specialty (yes, no). The reference category for each interaction was withholding and young age and specialty and young age, respectively. Ordinal regression models were also adjusted by gender, familiar with some IDD/ASD, religion, value assigned to the truth, participant institution (medical facility/university), knowledge about IDD, ASD, and ADHD, and bioethics courses. The differences were considered statistically significant when p <0.05. The Stata 14 software was used for all statistical analyses.