| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
School of Psychiatry and Behavioural Sciences (A-M B., B.T., F.C.), Rawnsley Building, Manchester Royal Infirmary, Manchester, UK and Department of Gastroenterology (Q.A.), Clinical Sciences Building, Hope Hospital, Salford, UK.
Address for correspondence Dr. Anne-Marie Biggs, Unit of Chronic Disease Epidemiology, The Medical School, University of Manchester, Oxford Road, Manchester, M13 9PT. E-mail: Anne-Marie.Biggs{at}man.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
MATERIALS AND METHODS: Consecutive new patients with upper abdominal or chest pain presenting to a secondary/tertiary clinic were assessed using the Childhood Experience of Care and Abuse and Life Events and Difficulties Schedules. They completed the Hospital Anxiety and Depression and Health Anxiety Questionnaires. Outcome was total number of health care visits recorded in hospital and general practice (GP) records over 18 months.
RESULTS: One hundred fifty-one patients were included (65% response rate). Health care visits were most frequent in unmarried (p < 0.0005), females (p < 0.0005), and those lacking social support (p = 0.012). In multiple regression analysis to predict number of health care visits, reported sexual abuse (p = 0.042) and death of a sibling during childhood (p = 0.026) were also independent predictors, together with SF36 subscale scores for physical function, health perception, and mental health (35% of variance explained). Childhood adversity predicted health care use in patients with functional gastrointestinal disorders and recent social stress did so in patients with demonstrated pathological findings.
CONCLUSION: After adjustment for demographic, physical, and psychological factors, childhood adversity, especially in severe form, is an independent predictor of health care use in patients with upper functional gastrointestinal disorders. The same was not true for patients consulting for demonstrable pathological abnormalities, for whom ongoing social stress was an independent predictor.
Key Words: health care use, • childhood adversity, • recent life adversity, • functional dyspepsia, • noncardiac chest pain, • gastroesophageal reflux disease, • ischemic heart disease.
Abbreviations: GERD = gastroesophageal reflux disease;; IHD = ischemic heart disease;; FD = functional dyspepsia;; NCCP = noncardiac chest pain.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
In addition, high health care utilization has been found to be associated with depression and anxiety in patients in whom there are demonstrable pathological findings, including myocardial infarction and angina (16,17), inflammatory bowel disease (18), and general medical symptoms (19–21). Therefore, any model testing psychosocial variables as predictors of high health care utilization should be applied to patients who consult for symptoms explained by demonstrable pathological findings as well as those with functional disorders.
Reported childhood neglect or abuse has been associated with high health care use in patients with any diagnoses (12–14,22,23), and two studies found a dose–response relationship between levels of abuse and use of health care (13,24). It is not clear whether this is a direct effect, however, because one of the few studies that has used an objective measure of health care use indicated that depression was more closely associated than sexual abuse (12). Consultation with dyspepsia is associated with the exposure to adverse life events in adulthood (11,25), but this has not been studied in patients with noncardiac chest pain, ischemic heart disease (IHD), and gastroesophageal reflux disease (GERD).
In patients with functional gastrointestinal disorders, Koloski (14) noted that very little of the variation in consultation behavior can be explained by symptom severity and called for more satisfactory studies of the influence of psychosocial variables so that a comprehensive model could be developed. Such studies require adequate measurement of childhood experiences, social stress and support, anxiety and depression, health anxiety and illness attitudes, and an objective record of consultation. The last is important because depression or health anxiety may distort memories about health care use. Our study was performed in the United Kingdom, where health care is provided free to the whole population through the National Health Service. This meant that comprehensive and accurate data regarding health care use at primary and secondary care could be determined from medical records.
In this study of patients presenting to secondary care with upper abdominal and chest pain, we assessed whether there was a direct association between the experience of childhood adversity or recent social stress on health care use once the effects of distress, health anxiety, and severity of current illness and symptoms had been taken into account and, if so, whether this was a specific or general aspect of childhood adversity. We tested the model first in all patients, then in patients with demonstrable pathological findings and functional gastrointestinal disorders separately. The specific hypotheses undergoing test were: (1) that the experience of early and/or recent adversity would be a significant predictor of health care use even after all other relevant disease factors have been accounted for; (2) that specific aspects of adversity (eg, sexual abuse) would be significant predictors of health care use even after other demographic, psychological, and illness factors had been accounted for; and (3) that the same predictors of health care use would hold in patients with a diagnosis of a functional gastrointestinal disorder and those with demonstrable pathological findings.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
The socio-demographic variables included age, sex, ethnicity, years of education, social class, and marital status. Each patient was asked to rate the severity of their pain, based on a scale of 0 (no pain) to 100 (most severe pain). The following instruments were administered at a semistructured interview in the patient’s home.
SF-36
The 36-item short-form health survey (SF-36) (27) includes eight scales to encompass health-related quality of life. Each component is scored from 1 to 100, with low scores indicating poor health. The questionnaire has been used among gastroenterology patients (18,28) and among cardiology patients (29).
HADS
The Hospital Anxiety and Depression Scale (HADS) (30) is designed to be administered in a medical clinic setting because it avoids physical items (eg, weight loss, pain) that might be caused by physical illness. It contains 14 items (7 anxiety and 7 depression); a total score for each is presented in this article. Recent reviews have shown that the questionnaire is acceptable to patients and is a valid and reliable tool in clinical populations (31,32).
IPQ
The Illness Perception Questionnaire (33,34) includes four subscales. The identity scale lists 12 bodily symptoms, which the respondent may attribute to the current disorder. The timeline scale assesses the respondent’s beliefs regarding the likely duration of the disorder, consequences scale assesses its likely effect on physical, social, and psychological functioning, and cure scale assesses individuals’ beliefs about cure or control of their condition. It has been used in many different patient groups.
HAQ
The Health Anxiety Questionnaire (35) focuses specifically on concerns about health, with questions about health worry and preoccupation, fear of illness and death, reassurance-seeking behavior, and interference with activities. High scores indicate more worry. The HAQ has been used among gastroenterology patients (15).
CECA
The Childhood Experience of Care and Abuse (CECA) interview was used to assess childhood adversity (36). It was developed as a retrospective, semistructured interview to collect and rate individuals’ experiences of childhood. The instrument uses ratings that are based on behavioral evidence and not influenced by respondents’ current mental state or reporting style. The rating of severity was made by an expert panel who were blind to patients’ clinical history and to whom the interviewer (A.M.B.) relayed the details of childhood experience. The original four-point scale was used to rate each domain on the CECA; subjects scoring 1 (marked) or 2 (moderate) were considered to have experienced childhood adversity and are described as such in this article. There are five core areas of childhood adversity. Antipathy covers parental dislike, criticism, and hostility toward the child. Neglect or indifference examines both physical aspects, such as inadequate clothing or food, and emotional aspects, such as lack of interest in the child’s well-being. Physical abuse takes into account the age of the child, the nature of the abuse, and the injury inflicted. Usually an implement such as a belt or stick is involved. Sexual abuse involves some physical contact. Psychological abuse includes incidents such as extreme rejection and calculated cruelty. Examples are given in an Appendix to illustrate the details elicited and the high thresholds used in the marked and moderate categories. Additional questions were added to the interview to ascertain whether any first degree relatives were seriously ill or died during the subject’s childhood. A similar degree of detail was elicited—ie, the nature of the illness, its severity and duration, and the subject’s age and other care/support at the time. Illness in either the subject or a first-degree relative was only included if it required substantial hospital treatment or was life-threatening. Deaths of a parent or a sibling were included if they occurred before the subject was aged 17. These variables were coded as yes or no. The variables were added because the CECA focuses only on the impact on care resulting from such circumstances. We were interested in whether exposure to childhood illness/death in the family affected the use of health services, irrespective of childhood abuse or neglect scores.
LEDS
The Life Events and Difficulties Schedule (37) gathers evidence-based descriptions of life events through a semistructured interview and assigns an objective rating. The questions cover areas such as health, housing, employment, finance, and relationships. The events and ongoing difficulties of the year preceding the clinic appointment formed the basis for the interview in the present study; difficulties involve ongoing problems such as living with a seriously ill family member or being in a violent marriage. Difficulties were scored from 1 to 7, and those rated 1, 2, or 3 were classed as severe; these scores were transformed into a scale that gave a high score for numerous and severe chronic social difficulties (38). Like the CECA interview, events and difficulties were rated by an expert panel blind to the clinical details. Events and difficulties which were associated with the abdominal or chest pain undergoing study, such as hospital admission and time off work, were excluded.
Health Care Use
Some months after completion of all of these assessments details of consultations with the general (family) practitioner, hospital specialists (outpatient visits), and with professions allied to medicine (eg, physiotherapist, occupational therapist, district nurse) were gathered from patients’ notes for 12 months before the index visit and 6 months after this date. The total number of visits over the 18 months was calculated and used as the main outcome variable in this article.
Statistical Method
The statistical package SPSS (version 10.1 for Windows 1998; SPSS) was used for all analyses. The total number of health care visits was not normally distributed but became so after log-transformation (using the one-sample Kolmogorov-Smirnov test). Univariate analyses compared each variable with total number of health care visits using the Mann-Whitney or Kruskal-Wallis tests. Results are presented with the median and interquartile range. Spearman’s correlation coefficient was used for continuous measures. Partial correlation was used to compare continuous variables while adjusting for sex, marital status, and availability of confidant.
Hierarchical stepwise multiple regression analysis with mean substitution was used to identify the factors most closely associated with health care use; log-transformed total number of health care visits was used as the dependent variable. In the first analysis the following were entered as independent variables in the first step: sex, single and separated/widowed/divorced status, years of education, access to confidant, diagnosis (as 4 dummy variables), pain score, eight SF-36 scores, HADS anxiety and depression scores, HAQ total score, and four IPQ scores. Recent social stress score, exposure to death or serious illness in the family during childhood, and any reported childhood adversity were entered in the second step.
The second analysis substituted for the last variables in the second step: the specific illness/death exposures (death of father, mother, or sibling, exposure to severe illness) and the six individual childhood adversities (antipathy from mother, antipathy from father, neglect, physical, psychological, and sexual abuse).
This analysis, entering all independent variables as a single step, was repeated separately for patients with demonstrable pathological findings (GERD and IHD) and those with functional gastrointestinal disorders (functional dyspepsia, noncardiac chest pain, and mixed diagnoses). The analysis was also repeated using number of visits to the general practitioner (GP) and other visits separately.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
Total Number of Health Care Visits
Over the 18-month period, the median number of visits was 14 (9 of which were at the GP). One quarter of the patients made 20 or more visits. There was a high correlation between total health care visits during the 12 months before and the 6 months after the index clinic visit (r = 0.56, p < 0.0005).
Demographic Details
Consultation rates were higher among women compared with men, among those who were single or separated/widowed/divorced compared with married, and among those who lacked or had limited access to a confidant (Table 1).
|
|
|
Childhood Adversity and Recent Social Stress
Increased health care use was associated with antipathy from father and with sexual abuse but not with other forms of childhood adversity. People who reported two or more childhood adversities showed the clearest increase in number of health care visits (Table 4, bottom row). This occurred only in patients with functional gastrointestinal disorders. Patients with demonstrable pathological abnormalities who reported none, one, or two childhood adversities had median numbers of 13, 11, and 18 health care visits, respectively (p = 0.56, Kruskal-Wallis). Corresponding values for patients with functional gastrointestinal disorders were 12, 17, and 20 (p = 0.029, Kruskal-Wallis).
|
The ongoing social difficulties score (excluding the subject’s own health) was associated with health care visits (r = 0.21, p = 0.011) after controlling for sex, marital status, and access to confidant.
Multivariate Analysis to Predict Health Care Use
In the first stepwise multiple regression analysis, the following were included in the final model to predict total number of (log-transformed) health care visits: SF36 scale scores for physical function, health perception, and mental health, sex, single status, and diagnosis of noncardiac chest pain; 30.4% of the variance was explained by this model. In the second analysis (Table 5) the same variables were included but, in addition, death of a sibling during childhood and reported sexual abuse were included; this model explained 35.4% of the variance (each of the additional variables led to a significant increase in R2).
|
|
The predictors of number of other visits (principally to hospital medical specialists and nonmedical health professionals in primary care) were: SF36 scores of physical function, mental health and health perception, sex, and severity of pain (25.4% variance explained).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
In relation to our first hypothesis, we found that childhood adversity and experience of a death of a sibling during childhood were directly associated with number of health care visits once the effects of demographic variables, diagnosis, health status, and illness perception had been accounted for in a multivariate analysis. In relation to our second hypothesis, we found that it was specific aspects of childhood experience (death of a sibling and sexual abuse) rather than a global measure of childhood adversity that were included in the final model of predictors of health care use. Thus, our data should be taken to indicate that some aspects of childhood adversity appear to have a lasting effect on the tendency to seek medical help. This is particularly clear in self-initiated visits to the GP and in patients with functional gastrointestinal disorders who reported two or more adversities. In the case of childhood death of a sibling, this might be a direct effect but it may be confined to a limited number of people; the effect was most apparent in multivariate analysis, in which other illness-related factors were taken into account. Very little previous work has considered the separate effects of family bereavement and abuse during childhood in relation to health care use. It is possible that both experiences sensitize individuals to the significance of bodily symptoms. Further work is needed to see if different mechanisms are involved.
Although they appeared as independent predictors in our analyses, psychological or sexual abuse were only reported by a small number of people, and nearly all of them reported at least one other form of childhood adversity. These forms of abuse may be markers of severe adversity or reflect particularly nasty aspects of family relationships. Previous reports found that frequent doctor visits were associated with sexual, as opposed to other forms, of abuse (13) or more severe abuse (24).
Childhood adversity made a modest independent contribution to our final model, in line with two previous reports (13,22). This is accounted for, in part, by the fact that childhood adversity was significantly associated with SF36 scores for physical function, health perception, and mental health, which may all be acting as mediators between childhood adversity and high health care use. Several mechanisms may be involved.
People who report childhood adversity have experienced disturbed close relationships and these may be reflected in an unsatisfactory doctor–patient relationship. The doctor’s attempts to allay the patient’s anxieties about health may not be successful in these conditions, leading to repeated consultations.
People with a history of childhood adversity and neglect may be less physically fit; they may be obese, smoke, drink heavily, lack regular exercise, and have more physician-coded physical diagnoses (40). This group is, therefore, at greater risk for IHD and other diseases and of more impaired function (the SF36 physical function scale measures ability to walk up stairs, etc). Childhood adversity has been associated also with a more negative view of health, even after the effect of depression has been controlled (41). This perception is likely to lead to more frequent consultations. Childhood adversity is also directly associated with poor mental health and depression and anxiety are both associated with increased health care use.
In addition, biological factors may be involved because childhood adversity has been shown to lead to greater sensitivity of the hypothalamic-pituitary-adrenal axis to stress in adulthood, making the individual more susceptible to pain and other symptoms, for which medical treatment may be sought (42).
Two other studies have used an objective measure of health care use. One (22) found that physical and emotional (but not sexual) abuse were related to increased doctor visits, but the other found that it was the combination of abuse with psychological distress, or distress alone, that was associated with high health care use (12). In our study, abuse was a predictor of health care use independent of distress. This may reflect the biological changes associated with childhood adversity mentioned. It may also reflect the altered perception of health in people with a history of childhood adversity. In our study childhood adversity was significantly associated with a more pessimistic view of the consequences of their illness (IPQ consequences score), even after diagnosis was controlled (data not shown). This effect was mediated by depression but also was associated with childhood adversity.
In relation to our third hypothesis, we found, first, that the overall number of health care visits was similar in patients with functional gastrointestinal disorders and those with demonstrable pathological findings. This corresponds with the finding of another United Kingdom study, in which approximately half of the most frequent attenders at hospital gastroenterology outpatient clinics had functional symptoms (39). Secondly, we found that the predictive models in these two groups were different. In the demonstrable pathological abnormalities group, the predictors of health care visits included the female predominance noted in previous literature (43) and single status. By contrast, demographic variables were not independent predictors among patients whose symptoms were those of functional gastrointestinal disorders (functional dyspepsia, noncardiac chest pain, and mixed groups), but reported childhood adversity (psychological abuse) was predictive of health care use. There seems to be a specific relationship between the reporting of two or more childhood adversities and high health care use only in the patients with functional gastrointestinal disorders group. In the demonstrable pathological abnormalities group ongoing social stress emerged as a predictor; the areas involved concerned serious problems in the domains of work, housing, financial, marital, and other relationship problems. This appears to be a relatively new finding, because few previous studies have assessed this relationship. Previous research demonstrating the relationship between previous stress and more frequent attendance at primary care did not distinguish diagnostic groups (44). It is likely that the frequency of health care in patients with demonstrable pathological abnormalities is dominated by the severity of the underlying pathology (eg, severity of IHD) as the SF36 physical function measure accounted for most of the variance (Table 6). Other psychological factors (eg, depression) either contribute to this measure or are explained by the demographic variables of female gender and single status, which are powerful predictors in this group. Other studies have tended to use ratings that are based on the subject’s assessment of stress, so judgment may be affected by current depression. In contrast, our study uses an interviewer-rated assessment.
The severity of pain was not clearly associated with health care use. This may be unexpected but it should be remembered that all the subjects were being seen in secondary care and the mean pain score (out of 100) was 58, ie, most patients had considerable pain. This may have played a part in the initial search for medical help but was not a characteristic which differentiated the intensity of health care use thereafter—other variables were more closely involved. Like others (25) we found that the degree of concern about the symptoms was more closely associated with health care use than actual pain severity. We did examine the association between pain severity and health care use; it was only significant for patients with dyspepsia, but numbers were small and this finding should be treated with caution.
Childhood factors were only important in predicting self-initiated appointments with the GP. In the United Kingdom, patients are free to make appointments with the GP on a daily basis, whereas hospital and other visits are usually arranged by the GP, who acts as a "gatekeeper" to specialist services (eg, hospital specialist or physiotherapy). Childhood experiences of adversity (antipathy from father during childhood) and death of a family member appear to be independent predictors of number of GP visits but not number of visits to other specialists. Sansone (22) also found that physical and emotional abuse, although associated with increased doctor visits, did not predict specialist referrals. Future research should consider more carefully the predictors of doctor visits that are initiated by the patient and those that are initiated by doctors.
Like others (14) we found that sex, marital status, and availability of a confidant were important influences on health care use; adjusting for these variables led to depression ceasing to be significantly associated with number of visits. It appears that depression itself is not the key variable in determining health care use—the number of bodily symptoms and worry about health and illness are more important (19,20). In this study we found that the IPQ and HAQ scale scores were significantly associated with health care use even after adjustments were made for sex, marital status, and availability of confidant. These did not appear in the final model, however, because they were highly correlated with the SF36 health perception subscale score. In fact, the SF36 subscale scores were remarkably predictive of health care use in both univariate and multivariate analyses.
Koloski et al. (14) proposed a model to explain frequent medical consultations for dyspepsia; this involved anxiety and depression, a history of physical or sexual abuse, high neuroticism, and poor social support. Our findings would appear to provide broad support for this multifactorial model in relation to patients with functional gastrointestinal disorders. We overcame the methodological shortcomings listed by Koloski et al. because we used detailed and observer-rated interviews for childhood experience and recent life stress, and we also used a precise, objective measure of health care use. In doing so, the findings with regard to social support are clear: people lacking social support, either defined as unmarried or lacking a close confidant, consult more frequently, whatever the nature of their illness. Once this effect has been taken into account, much of the variation can still be accounted for by severity of physical and psychological impairment, negative perception of health, and a reported history of abuse and family bereavement during childhood. This confirms the report of Bass (45) who found a high level of psychiatric disorder and reported abuse among frequent attenders at gastroenterology clinics with medically unexplained symptoms.
There were some limitations to our study. The numbers included were small in view of the many variables we included, especially as we divided the sample into demonstrable pathological abnormalities and functional gastrointestinal disorders groups. However, the number of subjects was limited by the detailed nature of the interviews, which we regard as essential for this type of study. We have presented results for the whole cohort, but we also felt it was necessary to examine our findings in relation to the nature of the underlying diagnosis and the type of medical visits; these must be regarded as preliminary findings and need to be repeated in larger samples. Our study was only partly prospective; ideally, we would have assessed people at the start of the 18 months, not after 12 months of seeking care in primary care. However, the results of the multiple regression were similar if we confined our analysis to health care use during the 6 months after the index consultation (data available on request). We included as many variables as we felt were reasonable, because a systematic review of frequent attendance in primary care (41) found that numerous dimensions were involved but no study had included all relevant ones. We had expected that the effect of sexual abuse on health care use would disappear once we controlled for health anxiety and illness beliefs. This proved not to be the case, which opens the way for further research examining more closely the possible mechanisms involved.
Interventions that might reduce the treatment seeking of this group of patients include cognitive behavioral therapy, psychodynamic interpersonal therapy, and antidepressants (46, 47). We have recently compared the last two in severe irritable bowel syndrome patients using a randomized, controlled, trial design (46). Both led to improved health-related quality of life at no additional cost. The psychotherapy, but not the antidepressants, led to a significant reduction of health care costs through reduced outpatient visits. It is probable that this form of therapy addresses the interpersonal problems that might have resulted from early childhood adversity. It is likely that the same parameters are related to health care use in irritable bowel syndrome as those we have recorded here in patients with upper functional gastrointestinal disorders. The beneficial effects of antidepressants are likely to be their effect in reducing depression, which contributes to the SF36 scores of mental health and health perception (46).
|
APPENDIX |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
Physical abuse rated 1: Male hit by older brother with stick (9–10 strokes) and punched on a daily basis from age 3 to 4 until he left home at age 13; also thrown against a door on several occasions. Never injured badly enough to go to hospital.
Physical abuse rated 2: Female age 4. Her father hit her at least once per week with the strap end of a belt. He was always drunk when he hit her; on one occasion at age 5, she was punched.
Physical abuse rated 3 (i.e., not included): Male, from age 6 to 17, was hit weekly or more on the back of the head by father; glancing blow, no mark left, not out of control, in context of being told off.
Sexual abuse rated 1: Male between the ages of 6 and 12 was raped by his brother almost every night. Unable to confide in anyone; his brother used physical violence to enforce his cooperation.
Sexual abuse rated 2: Female, recalls waking at age 3 because her older brother (4 years older than her) was trying to push a pen into her vagina. Until she was age 11, her brother attempted on several occasions (approximately 1 occasion per year) to have sex with her. "I didn’t want to and he hurt me."
Sexual abuse rated 3: (i.e., not included) Female age 10. She was cared for by her grandmother (see neglect rated 2 below). Uncles visited on the weekend. One got her into bed; she woke up with his hands on her chest. She moved away from him and he did not do anything else. He said he wanted to tell her about sex and "did I want to see?"
Psychological abuse rated 1: Female age 5, witnessed her brother age 4 drowning. Her mother blamed her for the death and refused to look after her. At age 15, she received a parcel of her brother’s bootees and other baby items and a note from her mother that read "I want you to keep these as a reminder of what you’ve done."
Neglect rated 2: Male who got himself up and ready for school from ages 5 to 6. Very few clothes. Mother went out to work at 7:00 PM, leaving the children in the house alone (he was oldest, age approximately 8). Poor diet that led to skin disorders, caused by lack of supervision of what he ate. No parental interest in school or job plans. Father uninvolved, spent all free time in pub. Occasionally saw father on weekends.
Antipathy rated 2 from father: Female with poor relationship with father, argued a lot, he was very unpredictable, never showed affection, no time spent together, never spoke to her in a pleasant manner, difficult to please, could not confide. Made frequent critical comments about her weight.
Antipathy rated 2 from father: Female with parents who showed no interest in school, friends, or her choice of career. No confiding. No recollection of events/days out/holidays. Birthdays remembered with small present.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication February 19, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONAPPENDIXACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. Moon, D. Blurton, and J. D. McCluskey General Strain Theory and Delinquency: Focusing on the Influences of Key Strain Characteristics on Delinquency Crime Delinquency, October 1, 2008; 54(4): 582 - 613. [Abstract] [PDF]
|
|
|
|
 |
|
 K. R. Gehrling and R. J. Memmott Adversity in the Context of the Neuman Systems Model Nurs Sci Q, April 1, 2008; 21(2): 135 - 137. [PDF]
|
|
|
|
 |
|
 R. C. Spiller Irritable bowel syndrome Br. Med. Bull., March 14, 2005; 72(1): 15 - 29. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Sheps, F. Creed, and R. E. Clouse Chest Pain in Patients With Cardiac and Noncardiac Disease Psychosom Med, November 1, 2004; 66(6): 861 - 867. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
