Modeling Daily Partner Violence and Substance Use Based upon Couple's Reporting

While agent-based models (ABMs) have successfully modeled violence and women's decision-making, they relied upon studies of her daily reports of violence and household environment; these models were not based upon descriptions of his emotions and perceptions. The purpose of this study was to improve our understanding of the triggers of violent events within violent relationships through agent-based modeling by including men's perceptions and emotions. An agent-based model was created of couples with history of violence based upon results of a study involving multiple time series of partner violence, including couples' daily reports of their emotions and perceptions. To explore factors that may alter model results, seven continuous variable parameters were created based upon significant (p ≤ .05) but discrepant (opposite directions) in prior studies. To assess the potential impact that influencing factors such as random stress as well as his and her feelings and behaviors could have on violence and stalking, the impact of these factors was also assessed. Results found that, at baseline, which included no extremes in variable parameters, no violence emerged. One prior-day→same-day relationship (HerConcern→HerConcern) was particularly important in this ABM. Men's and women's drug use and refraining from arguments had little impact on any outcome, but his and her alcohol use, his sense of insult and her violence all had significant effects. In fact, women's alcohol use interacted with other influencing variables and was a source of atypical patterns. In conclusion, incorporating men's perceptions into an ABM of partner violence resulted in important differences compared with ABMs based solely on women's input. Not only were women's daily concerns about the effect of violence on children was critical to results, but this ABM demonstrated the complexity of partner violence in response to influences as illustrated by contextual dependence, interaction effects and synergy.

Value of Modeling Violent Relationships

Intimate partner violence (IPV) is a scar on human society. Growing evidence demonstrates that relationships involving IPV and women's decisionmaking about violence represent complex phenomena, best studied as complex adaptive systems. Unfortunately, that complexity limits our ability to fully understand it. This paper presents findings from a series of agent-based models (ABMs) that were created from studies involving multiple time series of couples' daily reports of violence, perceptions and behaviors. To identify potential influencing variables, we modeled the impact that random stress and intentional behavior of the women could have on men's violence and stalking. ABM models of IPV noted the lack of violence at baseline without influence, and found that three variable parameters (Distance-Distance+, HerViolence-HerViolence+, Distance-Distance-) accounted for most patterns of violence development. Random stress and arguments had little effect and the nature of the alcoholviolence relationship remained unclear, however, her violence could increase his violence and stalking. One important difference between ABMs was the importance of the persistence of her concern about the effects of violence on children. Although only modeled in the second ABM, it proved critical to results. When modeling women's decision-making, her abstinence from arguments, alcohol use and violence had no effect on whether to seek help, take legal action or leave; random stress and her daily violence did not affect seeking counseling. However, daily arguments, forgiveness and heavy alcohol use did impact actiontaking, increasing counseling, legal action and leaving generally. The addition of catastrophe equations could alter these outcomes, resulting in more counseling but less legal action. In addition, children are very important when considering decision-making; concern for children affects violence while number of children affects decision-making. In conclusion, ABM can yield important insights into IPV and have clinical implications. It can provide greater understanding of the phenomenon and allow us to test the nature of correlations. (i.e., between alcohol use and violence). ABM can clarify the inherent complexity within violent couples and facilitate sense-making. Finally, it can allow clinicians to test interventions in vitro without risk to vulnerable women.

Same-Day Correlates and Prior-Day Predictors of Couples' Violent Behaviors Based upon Partners' Daily Reports

Although previous research identified predictors of violent events within violent heterosexual couples, findings were limited to the woman's reports, to her perceptions; his assessments were not obtained. This exploratory study was conducted to gain understanding of proximal predictors of violence assessed in "real-time" from the perspective of both partners. Fifteen adult heterosexual couples in which the woman reported experiencing partner violence in the prior 30 days were enrolled in a primary care clinic. Each partner provided separate daily telephone reports for eight weeks via an automated Interactive Voice Response (IVR), concerning the previous day's violence, alcohol use, stressors, emotional reactions and concerns for children. Same-day correlates were determined by Pearson correlations while prior-day predictors were identified via vector autoregression. Same-day correlations show that men's violence was associated with almost every other variable while women's violence correlated with men's violence, men's drug use, women's alcohol use, anger, closeness, hassles, and all men's negative feelings. Using prior-day predictors, men's violence was related to feelings (primarily hers), but women's violence was more dependent upon feelings of both of them as well as women's prior-day violence and alcohol use. Men's violence was dependent upon their partners' prior-day feelings and the men's lack of concern about effects of violence on children. Women's violence was also dependent upon women's prior-day feelings, as well as women's violence, alcohol use, marital closeness, and men's concern for children. Although the co-occurrence of men's and women's violence has been seen before, in this study only women's violence was linked to alcohol use.

Complex Relationship Between Daily Partner Violence and Alcohol Use Among Violent Heterosexual Men

Although alcohol use and partner violence are consistently associated, the nature of the alcohol-violence relationship is still unclear. The purpose of this pilot study was to use longitudinal daily assessments of male partners' alcohol use and violent events to identify the nature of the alcohol-violence relationship, employing both linear and nonlinear analyses. The participants were 20 adult heterosexual couples of whom the woman reported experiencing partner violence in the prior 30 days. Each partner provided a separate daily telephone report for 8 weeks via an automated interactive voice response (IVR), concerning the previous day's violence, alcohol use, stressors, emotional reactions, and concerns for children. Individual IVR databases were merged to form a combined couple's IVR time series. Time series were analyzed using graphic, linear, and nonlinear methods. Graphic analysis using state space grids found no consistent pattern across couples. Similarly, linear analysis using same-day cross-correlation and prior-day beta statistics found no significant group-level alcohol-violence relationship. Using cross-approximate entropy statistics and differential structural equation modeling, no nonlinear relationships between alcohol use and violence were noted either. Whether applying linear or nonlinear analytic methods, there is no group-level relationship between alcohol use by male perpetrators and their violent acts. The implications are significant. First, the alcohol-violence relationship may differ among subgroups. Second, couples need to be assessed thoroughly to determine their unique relationship with alcohol use, so that couple-specific interventions can be designed. Third, if perpetrators believe that their violence is facilitated by their alcohol use, then alcohol reduction should be encouraged despite any evidence suggesting a different alcohol-violence relationship. Finally, the accepted alcohol-causes-violence belief held by many providers needs to be reconsidered. Because the nature of the alcohol-violence relationship varies considerably across couples, clinicians should seek to understand their unique relationship applying across-the-board management approaches.

Violence towards women and their decisions to take action: A complex systems approach

Intimate partner violence (IPV) is a blight on society. Our traditional understanding suggests that interventions should be straightforward, leading to predictable positive results. However, these assumptions do not reflect the reality of IPV, which continues to frustrate physicians. While IPV research has thoroughly described the developmental risks and impacts of IPV, the violent incident itself remains largely unstudied and poorly understood. Although this lack of research may partially explain physician frustration and the limitations of our interventions, the greater problem may be our reliance upon the wrong paradigm in guiding our understanding. Complexity science says that systems are globally understandable, but not completely knowable. Our hypothesis is that IPV needs to be viewed as a complex adaptive system if we are to understand the phenomenon, identify expectations and appropriately intervene. When viewed through the lens of complexity science, IPV becomes less knowable and predictable, suggesting that interventionists should expect variable response. Research has indeed demonstrated that partner violence is a complex phenomenon with multiple, interdependent factors and a nonlinear trajectory. This nonlinearity/unpredictability can impact outcomes in IPV, often more so than the frequency or severity of the violence. Similarly, women's decision-making concerning the violence is also a nonlinear process dependent upon multiple factors and catastrophic influences. Once recognized, complexity science offers a novel approach to explain IPV's obfuscation and resistance to predictable change. Using the tempered expectations of a systems lens, the violent interdependencies can be clarified, the obscure causes of events can be visualized, and the temporal irregularities can be mapped. Not only can the disruptive tipping points, system feedforward propagations, powerful attractors and discontinuities compromise reasoned intervention, but these same factors, if understood, can be harnessed to foster and magnify circumstances that enable positive change.

Effect of Incorporating Catastrophic Equations into an Agent-Based Model of Women's Action-Taking in Violent Relationships

Based upon multiple time series studies, an agent-based model (ABM) of women's decision-making related to partner violence was developed. However, help-seeking, legal action and leaving could be modeled as catastrophic phenomena, but catastrophic effects were not included in this prior ABM. The purpose of this study was to incorporate cusp catastrophe equations into the prior ABM to determine the impact upon her action-taking. Building upon this prior ABM of women's decision-making, we added cusp-related equations and compared resultant patterns and impacts of stress and interventions against those of the prior ABM. To assess the potential impact that random stress and her behavior could have on violence and stalking, the effect of variable parameter settings of these factors were assessed. Adding cusp equations to the prior ABM resulted in more legal action-taking for most of these patterns, many showing the effects of the bifurcation variables, and more women leaving in half of the patterns, but at the expense of increasing instability and more distortion effects. However, help-seeking was most impacted by possible interventions. In conclu-sion, adding cusp equations to the model had significant effects on modeling action-taking, especially for legal action and leaving while affecting the impact of interventions on help-seeking.

Agent-Based Modeling of Day-to-Day Intimate Partner Violence

The purpose of this study was to develop a mathematical model of mutual partner violence and assess impact of her controllable behaviors on reducing violence. An agent-based model was created of couples with history of violence based upon results of two multiple time series studies of partner violence. To explore factors that may alter model results, eight continuous variable parameters were created based upon significant (p=.05) but discrepant (opposite directions) results from previous studies. To assess the potential impact that random stress and her behavior (arguments, forgiveness, alcohol use, violence) could have on violence and stalking, the impact of variable parameter settings of these factors were also assessed. The model identified 18 unique patterns were observed, grouped into five general categories. Added random stress contributed to his violence in only two patterns. Although avoiding participation in arguments had no effect, her forgiveness and elimination of alcohol use often reduced her violence only. However, consistent violence or nonviolence on her part sometimes affected his violence and stalking. In conclusion, while increasing forgiveness and reducing alcohol intake could reduce her violence, they generally had little effect on his. However, if she eliminated her violence, it could eliminate his violence and stalking in some situations.

Multi-day Patterns around Taking Action in Intimate Partner Violence

This study was to determine the multi-day relationships among stressors, need and taking action (seeking counseling, taking legal action, leaving) by women in violent relationships. Women with recent husband-to-wife abuse but not at high-risk for life-threatening violence were recruited from six primary care clinics and asked to complete a daily description of the previous day's violence, need-for-action, and stressors using Interactive Voice Response via telephone for 8 weeks. Taking action (seeking counseling, taking legal action, leaving) was determined via weekly telephone contacts. To identify day-to-day recurrent strings, we used orbital decomposition, limiting time series to 29 women who took action during the study. Multi-day patterns were not common in taking action and are unrelated to violence. Only one 5-day string involving seeking counseling was identified in which women felt a lack of control. While taking legal action was part of five 5-days strings, each string consisted of one day of taking legal action within four days of no need-for-action, violence or stalking. Finally, one 4-day string that began with leaving coupled with spouse's excessive alcohol intake but no violence was noted. Hence, decisions to take action in violent relationships are not typically multi-day decisions linked to violence, but rather sudden events triggered by loss of control, his alcohol intake or unique situational factors.

Measuring interdependence in ambulatory care

RATIONALE, AIMS AND OBJECTIVES

Complex systems differ from complicated systems in that they are nonlinear, unpredictable and lacking clear cause-and-effect relationships, largely due to the interdependence of their components (effects of interconnectedness on system behaviour and consequences). The purpose of this study was to demonstrate the potential for network density to serve as a measure of interdependence, assess its concurrent validity and test whether the use of valued or binary ties yields better results.

METHOD

This secondary analysis used the 2010 National Ambulatory Care Medical Survey to assess interdependence of 'top 20' diagnoses seen and medications prescribed for 14 specialties. The degree of interdependence was measured as the level of association between diagnoses and drug interactions among medications. Both valued and binary network densities were computed for each specialty. To assess concurrent validity, these measures were correlated with previously-derived valid measures of complexity of care using the same database, adjusting for diagnosis and medication diversity.

RESULTS

Partial correlations between diagnosis density, and both diagnosis and total input complexity, were significant, as were those between medication density and both medication and total output complexity; for both diagnosis and medication densities, adjusted correlations were higher for binary rather than valued densities.

CONCLUSION

This study demonstrated the feasibility and validity of using network density as a measure of interdependence. When adjusted for measure diversity, density-complexity correlations were significant and higher for binary than valued density. This approach complements other methods of estimating complexity of care and may be applicable to unique settings.

Effects of religious and spiritual variables on outcomes in violent relationships

OBJECTIVE

Religious and spiritual factors in intimate partner violence have received increasing attention. But are such factors related to outcomes in violent relationships? The purpose of this study was to assess the relative impact of spiritual symptoms and religious coping on attitudinal/behavioral and clinical outcomes among women in violent relationships.

METHODS

Adult women with a recent history of husband-to-wife physical abuse were recruited from six primary care clinics. Once enrolled, 200 subjects completed a baseline interview and daily assessment of level of violence, using the Interactive Verbal Response for 12 weeks. At the completion of the study, contact with each participant was attempted to determine whether she had either sought professional help or left the relationship. Three religious/spiritual variables were assessed at baseline-number of visits to a religious/spiritual counselor, religious coping, and severity of spiritual symptoms. Stepped multiple linear regression was used to explain factor-analyzed outcomes (coping and appraisals, hope and support, symptomatology, functional status, readiness for change, and medical utilization), adjusting for demographic, marital, childhood, mental health, and violence variables.

RESULTS

After controlling for duration, severity and dynamics of violence, the use of spiritual resources, and the level of spiritual symptoms were associated with most attitudinal/behavioral and clinical outcomes, while religious coping was only associated with staying in the relationship.

CONCLUSIONS

Religious and spiritual factors were associated with most outcomes. Spiritual symptoms had a consistently negative effect on outcomes while use of spiritual resources had variable effects. Religious coping was only associated with refraining from leaving the relationship.

Recurrent multi-day patterns of intimate partner violence and alcohol intake in violent relationships

Consistent links exist between male and female alcohol intake and intimate partner violence (IPV). However, the nature of the relationship remains unclear. This study explores the temporal relationships between violence and heavy alcohol intake, looking for multi-day patterns. 200 women with a recent history of husband-to-wife abuse from six primary care clinics were asked to complete daily assessments using Interactive Verbal Response (IVR) via telephone for 12 weeks. To identify recurrent strings of activities, we used orbital decomposition. Multi-day patterns were found at the 5-, 7- and 9-day levels, but most represented extensions of 4-day patterns. Overall, consecutive days of male-perpetrated, moderate-severe violence were common. In addition, heavy alcohol intake by the husband was underrepresented on days involving verbal abuse only but overrepresented in consecutive days of such abuse; husband's alcohol intake preceded his verbal abuse and a sequence of husband-perpetrated verbal abuse followed by mutual abuse followed by wife-perpetrated verbal abuse was noted. No patterns involved heavy alcohol intake by the wife. In conclusion, few patterns involved heavy alcohol intake by men and none by women. Although husband's heavy alcohol intake may contribute to onset and maintenance of verbal abuse, it plays little role in recurrent patterns of physical violence.

Multi-day recurrences of intimate partner violence and alcohol intake across dynamic patterns of violence

RATIONALE, AIMS AND OBJECTIVES

Consistent links exist between male alcohol intake and male-perpetrated intimate partner violence (IPV) as well as female alcohol intake and female-perpetrated IPV. However, the nature of the relationship remains unclear. This study attempted to identify unique alcohol-violence patterns within three different types of relationship dynamics to better understand the alcohol-violence relationship and its role in violence dynamics.

METHOD

Two hundred women in abusive relationships were recruited from six primary care clinics. Subjects completed daily assessments of their relationship using interactive verbal response via telephone for 12 weeks. Dynamic patterns (periodic, chaotic, random) were determined by positive versus negative Lyapunov exponents and measures of correlation dimension saturation. To identify recurrent day-to-day activities, we used orbital decomposition (based on symbolic dynamics).

RESULTS

Periodic dynamics included daily reports with mutual abuse and alcohol intake while random dynamics included a variety of patterns, especially those involving unequal mutual abuse. Unique strings for each dynamic pattern were examined. Periodic dynamics involved heavy alcohol intake by the husband or mutual moderate-severe violence. Random dynamics uniquely involved mutual verbal abuse with husband's alcohol intake on same or different days as well as husband-perpetrated moderate-severe violence with or without husband-perpetrated minor violence. Chaotic dynamics uniquely involved combinations from wife-perpetrated minor violence alone to combinations of husband's heavy alcohol intake (with or without husband-perpetrated minor violence), mutual verbal abuse, and husband-perpetrated verbal abuse (with or without husband's heavy alcohol intake).

CONCLUSION

Recurrent 4-day patterns were observed. Each dynamic pattern was characterized by recurrent strings unique to that pattern.

Webs of causation in violent relationships

RATIONALE, AIMS AND OBJECTIVES

Predictors of intimate partner violence (IPV) include husband, wife and relational characteristics. However, we know little about the proximal, day-to-day triggering of abusive events. The purpose of the study was to analyse the daily temporal relationships among environmental, relational and violence variables in violent marital relationships.

METHOD

Two hundred adult primary care women who experienced violence in the previous month were recruited from six primary care clinics. Women completed daily assessments of household environment, marital relationship and violence using telephone interactive verbal response for 12 weeks. Same-day correlates were sought using cross-correlations among the environmental, relational and violence factors. Prior-day and prior-week associations were sought using vector autoregressions.

RESULTS

Except for wife's alcohol intake, all household environment and relationship factors demonstrated significant same-day correlations with IPV. However, prior-day violence by the husband, hassles, lack of husband's alcohol intake, emotional upset and marital distance were significantly related to current husband-perpetrated violence. Wife's violence depended upon her prior-day violence and alcohol intake only. All factors were related to husband-perpetrated violence in the subsequent week. Only wife's alcohol intake and husband's seeking forgiveness did not feedforward. In addition to the presence of multiple interdependent factors, circular causality was noted for marital distance and feeling upset.

CONCLUSIONS

IPV was due to multiple interdependent factors, feedforward dynamics and circular causality as expected in complex systems. The complex dynamics imply that simple interventions may have little chance of success, but understanding couple-specific dynamics may allow women to recognize high-risk prior-day profiles and take preventive action.

Do violence dynamics matter?

RATIONALE, AIMS AND OBJECTIVES

Intimate partner violence is a complex, non-linear phenomenon. The purpose of this study was to determine whether violence dynamics (pattern, degree of non-linearity, optimal non-linearity) contributed to outcomes in violent relationships.

METHODS

The study was conducted in six primary care clinics, enrolling 200 adult women in violent relationships. In addition to baseline and end-of-study interviews, women completed daily telephone assessments of household environment and partner violence using interactive verbal response. Three non-linearity measures of violence were computed with 'optimal' non-linearity estimated using Z-transformations. Assignment of dynamic patterns (periodic, chaotic, random) was made based upon Lyapunov exponent and correlation dimension. Outcomes across dynamic patterns were analysed using analysis of variance. In addition, stepped multiple linear regression explained factor-analysed outcomes, adjusting for demographic, childhood, mental health and marital variables; attitudinal/behavioural outcomes were also adjusted for when explaining clinical outcomes.

RESULTS

Women experiencing periodic violence recognized the importance of violence and used their active coping to seek mental health care. Those with chaotic dynamics recognized that they were not responsible, experienced fewer psychological symptoms and emotional role limitations, and did not seek help. Those experiencing random violence recognized its unpredictability and uncontrollability. Violence non-linearity predicted negative coping, positive appraisals and hope/support in regression analyses, while optimal non-linearity contributed to readiness for change and symptoms functioning. Of the nine outcomes investigated, violence non-linearity contributed to five outcomes.

CONCLUSION

Dynamic pattern of violence, degree of violence non-linearity and optimal non-linearity correlated with several attitudinal/behavioural and clinical outcomes. Knowledge of violence dynamics may have applications when working with violent couples.

Dynamics of violence

RATIONALE, AIMS AND OBJECTIVES

Three behavioural models suggest different dynamic patterns of intimate partner violence (IPV). However, few studies permit assessment of IPV dynamics. The purpose of this study was to estimate the degree of non-linearity in daily violence between partners over a 3-month period, identify their specific dynamic patterns and determine whether measures of violence severity and dynamics are interrelated.

METHODS

From six primary care clinics, we enrolled 200 adult women who experienced violence in the previous month and asked them to complete daily telephone assessments of household environment, marital relationship and violence using Interactive Verbal Response. To assess non-linearity of violence, algorithmic complexity was measured by LZ complexity and lack of regularity was measured by approximate entropy. Lyapunov exponents and correlation dimension saturation were used to approximate dynamic patterns.

RESULTS

Of the 9618 daily reports, women reported experiencing abuse on 39% of days, while perpetrating violence themselves on 23% of days. Most (59%) displayed random dynamics, 30% showed chaotic and 12% showed periodic dynamics. All three measures of non-linearity consistently demonstrated non-linear patterns of violence. Using multivariate analysis of variance, neither episode severity for men or women showed significant differences across dynamic types, but chaotic dynamics had the lowest frequencies of violence in men and women while random dynamics had the highest frequencies. Approximate entropy was positively correlated with violence frequency and burden in men and women, but Lyapunov exponent was inversely related to violence. LZ complexity correlated positively with wife-perpetrated violence only.

CONCLUSIONS

IPV is rarely a predictable, periodic phenomenon; no behavioural model describes the violence dynamics for all violent relationships. Yet, the measures of non-linearity and specific dynamic patterns correlate with different violent features of these relationships.

Bifurcating effects of hope and support in short- and long-term health outcomes among primary care patients without mental illness

RATIONALE

Adverse life events and stressors can lead to symptoms, increased alcohol intake, and impaired functional status, while hope and social support can mitigate such adverse effects. Yet, there is reason to question such simple, linear relationships among healthy adults. The purpose of this study was to assess whether short-term or long-term changes in alcohol intake, psychological symptoms or functional status were better modelled as cusp catastrophic or linear processes among primary care patients without mental illness.

METHODS

This secondary analysis of a study on the stability of psychological symptoms among primary care patients without mental disorders included 38 subjects who completed baseline, and 2-month and/or 6-month assessments of psychological symptoms, functional status and stressors as well as hope and social support. The analyses modelled short-term and long-term changes in alcohol intake, psychological symptoms and functional status using cusp catastrophe (CCM) and linear modelling.

RESULTS

Overall, four of the nine 2-month analyses found CCM superior to linear models; however, only one 6-month analysis (alcohol intake) found that CCM was superior. The 2-month cusp phenomena included both symptomatology and functional status. The asymmetry variables were often not significant in the CCM models; in fact, only distress was significant at all. While hope was a significant bifurcation variable at both the 2-month and 6-month levels, social support was a significant bifurcation variable for three of the four 2-month CCMs.

CONCLUSION

In conclusion, while 6-month outcomes were rarely explained through CCM, half of 2-month outcomes were. Hope and support demonstrated bifurcation effects.

Effects of family presence on the content and dynamics of the clinical encounter among diabetic patients

RATIONALE, AIMS AND OBJECTIVES

Primary care visits often include a family member or friend. The purpose of this study was to determine the effect of the presence of a family member on the visit content and dynamics among diabetic patients in Family Medicine settings.

METHOD

Encounters of patients with type 2 diabetes from 20 primary care clinics were audio-recorded and transcribed. Encounters were coded using the Davis Observation Codes, classifying content into 20 different categories at 15-second intervals. A random sample of 30 patients with family members was selected; 30 encounters in which no family was present were then matched to the randomly selected patients so that they would be similar group-wise in A1C level, length of visit, level of distress and discussion of non-patient family problems for analysis using orbital decomposition, an analytic technique based on symbolic dynamics in which categorical time series data are used to identify amount of complexity present and recurrent patterns of strings.

RESULTS

Visits were more linear if family members were present. When family members were present, 90-second strings of preventive services and evaluation/feedback were observed while 90 seconds of exercise discussion occurred when they were absent. Visits without family members tended to include more chatting, compliance discussion and nutrition counselling, while those with family members included more patient questions and evaluation/feedback. Finally, the sequence of history-to-planning-to-evaluation was observed when family were absent, but evaluation-to-planning-to-history when family were present.

CONCLUSION

The presence of a family member was associated with increased linearity and recurrent patterns that focused more on evaluation/feedback, preventive services, and patient questions, and less on chatting, exercise, compliance and nutrition in diabetic encounters.

Differences in social network structure and support among women in violent relationships

Social support is critical for women in abusive relationships. While social support may decrease the risk of getting into a violent relationship, it can also allow battered women to remain in violent relationships by reducing the negative impact of intimate partner violence (IPV). In addition to the social isolation that women in abusive relationships may experience, her social contacts appear to be important. The purpose of this study was to compare the size, structure, and composition of the social networks of women in abusive relationships with those of a matched cohort of nonabused women. The authors enrolled women from primary care clinics who reported abuse within the prior month, and a demographically matched comparison group of women in nonabusive relationships. Participants completed a social network analysis, and investigators compared social networks of abused with nonabused women. The networks of women in abusive relationships were smaller in size, but more efficient in their ability to reach their members, than those of nonabused women. Proportionally, networks of abused women had more women but fewer in-laws than those of comparison women. The women in these abusive relationships had higher measures of centrality, suggesting that they were more critical in holding their networks together. Yet, they had fewer social contacts, and provided more support than they received with fewer reciprocated ties. Thus, social networks of women in violent relationships are small and offer less support than those of comparison women. However, previous work on networks with weak ties and structural holes suggests that access to resources may be available through these networks.

Co-evolution of departmental research collaboration and scholarly outcomes

RATIONALE

Understanding scientific collaboration networks can assist research centres to develop strategies to maximize productivity, and help diagnose the causes of low system productivity. The purpose of this study was to use social network analysis to better understand how research collaboration within a productive department impacts scholarly productivity individually and departmentally.

METHODS

Over a 13-year period, departmental faculty completed an annual survey describing their research collaborations and scholarly productivity. Data were analyzed using social network analysis. Quadratic assignment procedure regression assessed the predictive value that an individual's measures of centrality within the network and effective size of their own network (egonet) had each year in predicting each scholarly outcome. Simulation Investigation for Empirical Network Analysis software assessed the co-evolution of the collaborative network and scholarship.

RESULTS

While no consistent patterns for individual's presentations were seen, individual's publications were associated with betweenness and eigenvector centrality, and effective egonet size. Grant submissions were associated with degree and eigenvector centrality, as well as effective egonet size. Departmentally, network dynamics depended upon the scholarship of those around you, but none of the forms of scholarship depended upon network characteristics. Of the three forms of scholarship, network dynamics depended primarily on publications in others.

CONCLUSION

Although individual scholarship was dependent upon individual centrality and effective egonet size, research collaboration within the department depended upon reciprocity, transitivity and scholarly productivity of its constituent investigators. Scholarly dynamics, at a departmental level, did not depend upon network characteristics.

Effect of comorbid depression on outcomes in diabetes and its relationship to quality of care and patient adherence: a statewide primary care ambulatory research and resources consortium study

OBJECTIVE

To determine whether current depression was associated with poorer quality of care and poorer patient adherence to treatment regimens and whether current depression was associated with patient diabetes outcomes independent of its relationships to quality of care and patient adherence among patients with diabetes.

METHOD

This study was conducted in the offices of family physicians who belong to the Statewide Primary Care Ambulatory Research and Resources Consortium from March 2006 to March 2011. Seven primary care physicians enrolled 10 to 20 English- or Spanish-speaking patients with diabetes presenting for routine follow-up visits. Subjects included 106 patients who completed a questionnaire documenting their depressive symptoms, compliance with diabetes therapy, diabetes-related quality of life, and patient satisfaction. The physicians completed a 4-item questionnaire concerning whether the patient had depression and any depression treatments that they ordered. All questions were answered either "yes" or "no." A practice research coordinator evaluated the quality of diabetes care provided and ordered hemoglobin A(1c) (HbA(1c)) testing for the patient.

RESULTS

Depression was associated with poorer compliance, quality of care, diabetes-related quality of life, and patient satisfaction; only HbA(1c) levels did not correlate with depression. When adjusting for compliance and quality of care, depression was still associated with poorer quality of life and satisfaction (P ≤ .001). While physician recognition and treatment of depression were less than optimal, depression severity was a significant predictor of receiving some form of mental health intervention (P ≤ .05) except for the provision of mental health counseling. Poor diabetes control was associated with the provision of counseling (P ≤ .10), while poor quality of life was associated with recognition of depression (P ≤ .10).

CONCLUSION

Depression was independently associated with satisfaction and quality of life but not diabetes control. Although depression severity was an important predictor of depression recognition and treatment, poor quality of life was a predictor of recognition, and poor diabetes control was a predictor of receiving mental health counseling.

Evolution of the research collaboration network in a productive department

RATIONALE

Understanding collaboration networks can facilitate the research growth of new or developing departments. The purpose of this study was to use social network analysis to understand how the research collaboration network evolved within a productive department.

METHODS

Over a 13-year period, a departmental faculty completed an annual survey describing their research collaborations. Data were analyzed using social network analysis. Network measures focused on connectedness, distance, groupings and heterogeneity of distribution, while measures for the research director and external collaboration focused on centrality and roles within the network. Longitudinal patterns of network collaboration were assessed using Simulation Investigation for Empirical Network Analysis software (University of Groningen, Groningen, Netherlands).

RESULTS

Based upon the number of active research projects, research development can be divided into three phases. The initial development phase was characterized by increasing centralization and collaboration focused within a single subject area. During the maintenance phase, measures went through cycles, possibly because of changes in faculty composition. While the research director was not a 'key player' within the network during the first several years, external collaboration played a central role during all phases. Longitudinal analysis found that forming ties was more likely when the opportunity for network closure existed and when those around you are principal investigators (PIs).

CONCLUSION

Initial development of research relied heavily upon a centralized network involving external collaboration; a central position of the research director during research development was not important. Changes in collaboration depended upon faculty gender and tenure track as well as transitivity and the 'popularity of PIs'.

Dynamic patterns in mood among newly diagnosed patients with major depressive episode or panic disorder and normal controls

OBJECTIVE

The purpose of this pilot study was to compare the dynamic patterns of hourly mood variation among newly diagnosed primary care patients with major depressive disorder or panic disorder with patterns in patients with neither disorder.

METHOD

Five adult patients with major depressive episode, 5 with panic disorder, and 5 with neither disorder were asked to complete hourly self-assessments of anxiety and depression (using 100-mm visual analog scales) for each hour they were awake during a 30-day period. Time series were analyzed using ARIMA (autoregression, integration, moving average) modeling (to assess periodicity), Lyapunov exponents (to assess sensitivity to initial conditions indicative of chaotic patterns), and correlation dimension saturation (to assess whether an attractor is limiting change). The study was conducted from March to June 2003.

RESULTS

Controls displayed circadian rhythms with underlying chaotic variability. Depressed patients did not display circadian rhythm, but did show chaotic dynamics. Panic disorder patients showed circadian rhythms, but 2 of the 4 patients completing the self-assessments displayed nonchaotic underlying patterns.

CONCLUSIONS

Patients with major depressive disorder or panic disorder may differ from controls and from each other in their patterns of mood variability. There is a need for more research on the dynamics of mood among patients with mental disorders.

Is research publication a catastrophic phenomenon among medical faculty?

Studies seeking to predict publication rates among faculty have found contradictory results. The purpose of this study was to determine whether short- or long-term research publications among family medicine faculty were better accounted for using cusp catastrophe modeling (CCM) rather than linear modeling. This secondary analysis of annual research publications used data collected from family medicine faculty in a university department. To predict the number of research publications, two service variables -- national service and administrative responsibility -- were used. There were three bifurcation variables: Scholarly Activity, Professional Status, and 'proportion of studies asprincipal investigator'. Research publications at two and five years were modeled using CCM as well as two linear models. Based upon the amount of variance explained, while linear models accounted for more variance in publications at -year intervals, CCM was superior at explaining publications for all three bifurcation variables at -year intervals. Entering all of the bifurcation variables into the models found that CCM explained more of the -year publication variance with Scholarly Activity and national service as significant predictors. In conclusion, short-term career planning needs to consider its irregular cusp behavior and to minimize the possible impact of bifurcation factors.

Trends in the perceived complexity of primary health care: a secondary analysis

RATIONALE

If the complexity of the patient's medical problems increases or the complexity of the interactions between the doctor and the patient, the staff or the health care system increase, then complexity of patient care will increase. This study examined trends in patient complexity, and identified doctor, practice and improvement strategy characteristics associated with perceived complexity.

METHODS

This secondary analysis used data from three Community Tracking Surveys with 22,134 primary care doctors completing surveys about themselves, their practice setting, practice improvement strategies and complexity of care in three consecutive 2-year time periods (1996-1997, 1998-1999, 2000-2001). Data were analysed using hierarchical logistic regression.

RESULTS

The proportion of primary care doctors who perceived that complexity of care had increased over the past 2 years rose from 31.5% to 35.9%. Perceived complexity of patient care was consistently related to being in solo practice and the belief that they could not frequently obtain high-quality services and referrals for patients. As availability of services increased, complexity decreased whereas as use of practice improvement strategies increased, complexity also increased.

CONCLUSIONS

Understanding that we cannot determine whether respondents understood care as 'complicated' or 'complex', potential consequences of this increase in complexity include an increase in medical errors and referral rates along with decreased quality of patient care and career satisfaction.

Recurrent patterns of daily intimate partner violence and environment

Although predictors of violent relationships have been identified, we are only beginning to understand the day-to-day dynamics of domestic violence. The objective of this study was to identify commonly seen patterns and strings of consecutive days involving verbal or physical abuse, and their preceding and subsequent events. Adult women (n=20) seen in a primary care clinic who experienced violence within the past month were enrolled. Subjects completed a daily telephone assessment of household environment and marital relationship for two months using Interactive Verbal Response (IVR). Results were analyzed using orbital decomposition, an analytic technique based on symbolic dynamics, in which categorical time series data are used to identify recurrent patterns of strings and quantify their complexity. While days without abuse had varied patterns involving arguments, stress levels, daily hassles, husband's alcohol intake, and sense of closeness (27 unique patterns), days involving verbal or physical abuse included a narrower range of patterns (15 patterns for verbal and 16 patterns for physical abuse). Daily patterns appear to cluster in triplets (3 consecutive days) of activity and show nonlinearity with triplets involving verbal abuse and triplets involving physical violence. Triplets involving either verbal or physical abuse were associated with arguments and high stress, but differed in the consistency of association with hassles, alcohol intake, and closeness. Finally, physical and verbal abuse tended to self-propagate. However, days involving verbal abuse did not precede days involving physical violence. In conclusions, while patterns of violence and household environments followed a nonlinear trajectory, only a limited set of patterns were observed. Although violence led to more violence, verbal abuse did not necessarily lead to physical aggression. In fact, verbal abuse and physical violence differed in the consistency of their relationships to hassles, husband's alcohol intake, and closeness.

Power law relationships between health care utilization and symptom assessment among people with panic attacks

RATIONALE

Although people with panic attacks are high utilizers of health care, the role of symptom assessment in care-seeking is unclear. Previous studies suggest that symptom perceptions are linearly related to utilization but panic appraisal is not. The purpose of this study was to determine whether the relationships between symptom assessment and utilization are non-linear, displaying power law distributions.

METHODS

This community-based study of 97 subjects with panic attacks assessed utilization of family doctor offices, total ambulatory utilization, and hospitalizations as well as symptom perceptions and panic appraisals. Matrices of symptom assessment versus utilization were created, and log-log plots were constructed. To minimize the risk of overestimation of power law distributions, linear, quadratic and cubic regression models were computed.

RESULTS

None of the utilization versus symptom perceptions displayed power law distributions. However, all three measures of utilization showed power law relationships with panic appraisals, but in unique patterns.

CONCLUSIONS

Although power law relationships were not found between symptom perceptions and utilization, unique patterns of power laws were identified between panic appraisals and all three measures of utilization.

Dynamical difference in patients encounters involving uncontrolled diabetes: an orbital decomposition analysis

RATIONALE

Poor glucose control is common in patients with type 2 diabetes. Little is known about the dynamics within the doctor-patient encounter that might explain this phenomenon. The purpose of this study was to compare dynamics of encounters with and without a hypoglycaemic medication change for patients with poorly controlled diabetes.

METHODS

The doctor-patient encounters of 182 patients with type 2 diabetes from 20 primary care clinics were audio-recorded and transcribed. Encounters were coded using the Davis Observation Codes (DOCs), classifying content into 20 different categories, for example, chatting or history taking, at 15 second intervals. Of the 60 encounters in which the A1C > 8.0, 25 involved a medication change. Fifteen patients were randomly selected from those with a change in medication as well as fifteen patients from those without a change in medication for analysis using orbital decomposition. 'Orbital decomposition' is an analytic technique based on symbolic dynamics in which categorical time series data, such as a string of DOCs, are used to identify amount of complexity present and recurrent patterns of strings.

RESULTS

Encounters with a change were longer (mean 20 versus 15.5 minutes) and included more time planning treatment (29% versus 23%). Encounters with and without a change displayed similar degrees of non-linearity, but change encounters were slightly more non-linear (D(Lyapunov) = 1.94 versus 1.75). Encounters with a change had more structure to them: they had many more DOC strings (60 versus 33 strings occurring at least three times), and those DOC strings more often linked treatment planning to history taking, chatting, health education, physical examination and compliance assessment. Encounters without a change linked treatment planning to history taking, compliance assessment and nutrition counselling but had no strings with chatting or evaluation and feedback.

CONCLUSION

In conclusion, the doctor's decision to change medication may be made before the encounter, thus the different strings of DOC codes observed. Chatting and evaluation and feedback may be strategies to increase trust before recommending a medication change.

Explaining health care utilization for panic attacks using cusp catastrophe modeling

Despite increased health care utilization, patients with panic disorder continue to report unmet needs. The objective was to compare the fit of linear and Cusp Catastrophe Modeling in explaining changes in utilization of emergency, general and mental health settings, and self-treatments for panic symptoms. This community-based study surveyed 97 subjects with panic attacks drawn from a sample of randomly-selected adults from randomly-selected households. The stressor (splitting) variable used was Phobic Anxiety while predisposing variables included Family Health Care Utilization, Perceived Life Threat and Need For Treatment, and Treatment Experience. Outcomes consisted of the number of sites and self-treatments used for panic symptoms when first seeking care and during the 2 months prior to survey. Use of mental health sites and self-treatments demonstrated superior modeling with cusp catastrophe approaches using treatment experience as the predisposing variable, accounting for 47% and 38% of variances respectively, improving the fit by over 20% compared to the best linear models in both cases. Cusp catastrophe modeling accounted for more variance than all linear models when describing use of mental health settings and self-treatments. Cusp catastrophe may explain bimodal distributions in behavior, delays in behavior change, and sudden shifts in behavior in stressful situations.

Assessing the biopsychosociospiritual model in primary care: development of the biopsychosociospiritual inventory (BioPSSI)

OBJECTIVE

There is growing evidence in support of the Biopsychosociospiritual Model. However, a cohesive instrument to measure each dimension in terms of its dimension-specific symptoms and functional status does not exist, serving as an obstacle to future research in this area. The purpose of this study was to develop and validate an instrument to measure each dimension in terms of its dimension-specific symptoms, appraisals, and functional status in an unselected group of primary care patients.

METHOD

An instrument that assessed biopsychosociospiritual symptoms and function and health outcomes was administered to 289 patients attending two primary care clinics. Responses were analyzed using principal component factor analysis with Equimax rotation. This led to the development of five Biopsychosociospiritual Inventory scales (impaired functional status, physical symptoms, psychological symptoms, social symptoms, and spiritual symptoms). Demographic differences in mean scale scores were sought. In addition to internal consistency, construct validity was assessed based upon dimension-specific health care utilization, life satisfaction, and perceived health status.

RESULTS

All five scales had excellent internal consistency (alpha > 0.8) and construct validity. Differences were strongly related to income, marital status, and employment in a manner consistent with previous research.

CONCLUSIONS

This study developed and validated the Biopsychosociospiritual Inventory which could potentially provide a holistic estimate of the impact of disease and its treatment, support research in this area, and lead to the expansion of classification systems that include spirituality. Further validation of this instrument in other primary care sites using diverse patient populations as well as its function over time is needed.

Dynamic covariation of symptoms of anxiety and depression among newly-diagnosed patients with major depressive episode, panic disorder, and controls

The purpose of this study was to explore the relationship between the dynamics of symptoms of anxiety and depression in primary care patients with newly-diagnosed major depressive episode (MDE), panic disorder, and neither disorder. Five adult patients presenting each with MDE, panic disorder, and neither disorder were recruited from two primary care clinics. Hourly self-assessments of levels of anxiety and depression were made for a 30-day period. Using differential structural equation modeling analysis, MDE and panic disorder differed from controls, but not from each other, in symptom levels, covariability, and effect on depression acceleration. Using state space grid analysis, attractors in controls were larger, more "influential", and more consistent. Patients with MDE had more total attractors but less consistent Week 1 attractors than patients with panic disorder or controls, suggesting more transient and less stable dynamics. This exploratory study suggests the importance of looking at covariability of anxiety and depression in mental illness.

Creating innovative research designs: the 10-year Methodological Think Tank case study

PURPOSE

Addressing important but complex research questions often necessitates the creation of innovative mixed methods designs. This report describes an approach to developing research designs for studying important but methodologically challenging research questions.

METHODS

The Methodological Think Tank has been held annually in conjunction with the Primary Care Research Methods and Statistics Conference in San Antonio since 1994. A group of 3 to 4 methodologists with expertise balanced between quantitative and qualitative backgrounds is invited by the think tank coordinators to serve on a 2-day think tank to discuss a research question selected from those submitted in response to a call for proposals. During the first half-day, these experts explore the content area with the investigator, often challenging beliefs and assumptions. During the second half-day, the think tank participants systematically prune potential approaches until a desirable research method is identified.

RESULTS

To date, the most recent 7 think tanks have produced fundable research designs, with 1 being funded by a K award and 4 by R01 grants. All participating investigators attributed much of their success to think tank participation. Lessons learned include (1) the importance of careful selection of participating methodologists, (2) all think tank communities of inquiry must go through 4 stages of development from pseudocommunity to community, and (3) the critical importance of listening by the investigator.

CONCLUSION

Researchers and academic departments could use this process locally to develop innovative research designs.

Factors that predict how women label their own childhood sexual abuse

Despite the psychological impact of child sexual abuse, many victims do not acknowledge that their experiences were "abuse". This study sought to identify factors that predict how women label their own experiences of childhood sexual abuse. This cross-sectional study was conducted in a family medicine clinic with adult female patients. Subjects completed structured interviews about their childhood environment and their sexual abuse history. Logistic regression analysis showed that labeling of abuse was dependent upon intercourse (beta = 7.43, p = .006), the frequency of abuse by the first perpetrator (beta = 5.08, p = .024), and paternal overprotection (beta = 6.69, p = .010). Findings suggest that the severity of abusive acts is most important and an over-protective father may enhance the victim's acknowledgment that sexual touching is abusive.

Predictors of development of adult psychopathology in female victims of childhood sexual abuse

The purpose of this study is to identify predictors of resilience and adult mental disorders in women with a history of childhood sexual abuse. This cross-sectional study was conducted in a family practice center using adult female (age 18-40) patients. Outcome measures assessed the prevalence of major depressive episode, panic disorder, agoraphobia, substance abuse, posttraumatic stress disorder, borderline personality disorder, bulimia, and suicidality. Seventy-six percent of the 90 women with sufficient data met criteria for at least one adult disorder. Mental health was related to high SES, lack of family alcohol abuse, lower frequency of first perpetrator abuse, and few perpetrators. Specifics of the abuse were associated with development of borderline personality disorder, substance abuse, major depressive episode, suicidality, bulimia, agoraphobia, and panic disorder. Maternal violence against the father, substance abuse within the household of origin, and maternal care and overprotection were also important. The specifics about the abuse and the family environment during childhood are important predictors of adult psychopathology.

Knowledge About Recommended Treatment and Management of Major Depressive Disorder, Panic Disorder, and Generalized Anxiety Disorder Among Family Physicians

BACKGROUND: Concerns have been raised about whether primary care physicians appropriately manage mental disorders. We assessed family physicians' knowledge of appropriate management of major depressive disorder (MDD), panic disorder, and generalized anxiety disorder (GAD). METHOD: Active members of the Texas Academy of Family Physicians (N = 3553) were mailed a questionnaire in 2002 asking them to indicate which treatments they felt were effective for MDD, panic disorder, and GAD and also to indicate how they had treated their last patient with each disorder. Their treatment strategies were then compared with current guidelines. RESULTS: 574 physicians (16%) responded. The percentage of respondents scoring at or above 80% for knowledge of effective treatments was 88.3% for MDD, 16.8% for panic disorder, and 12.5% for GAD (p <.001 for MDD vs. panic disorder or GAD). Only 0.3% of MDD patients, 1.4% of panic disorder patients, and 4.0% of GAD patients were not prescribed at least 1 of the effective treatments. Referral rates to mental health providers were high for all 3 conditions. CONCLUSIONS: There were significant gaps in physician knowledge of current guidelines on treating panic disorder and GAD, but not MDD. However, most patients with one of the disorders were either referred to a mental health provider or treated with an effective modality.

Panic plaques: panic disorder & coronary artery disease in patients with chest pain

BACKGROUND

The purpose of this systematic review was to identify characteristics of the chest pain associated with the presence of panic disorder, to determine the strength of the association between panic disorder and coronary artery disease (CAD), and to determine the association between panic disorder and known cardiovascular risk factors.

METHODS

Potential studies were identified via computerized search using MEDLINE and PSYCINFO databases, and review of bibliographies. MeSH headings used included "panic disorder" with "chest pain," "panic disorder" with "coronary disease or cardiovascular disorders or heart disorders," and "panic disorder" with "cholesterol or essential hypertension or tobacco smoking." Studies had to base their diagnosis of panic disorder on criteria from the Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, and objective criteria of CAD and risk factors had to be used. Only case-control and cohort studies were included.

RESULTS

The relative risk of panic disorder in patients with nonanginal chest pain is 2.03 [confidence interval (CI), 1.41 to 2.92]. Concerning the relationship between panic disorder and CAD, studies conducted in emergency departments found a relative risk of 1.25 (CI, 0.87 to 1.80). However, there is an inverse relationship between the prevalence of CAD in the study and the prevalence of panic disorder among the patients with CAD (r = -.469, P =.086). Panic disorder has also been linked to cardiac risk factors.

CONCLUSIONS

Panic disorder and CAD are correlated in noncardiology settings, and recurrent panic attacks may actually cause CAD. Recognition of either condition should lead the family physician to consider the other, resulting in increased vigilance and possible screening.

A Randomized Trial of the Effects of Remote Intercessory Prayer: Interactions with Personal Beliefs on Problem-Specific Outcomes and Functional Status

OBJECTIVES

Investigate the relevance of interpersonal belief factors as modifiers of the effectiveness of intercessory prayer.

DESIGN

Randomized clinical trial.

SETTING/LOCATION

Community-dwelling adults recruited from seven local church groups.

SUBJECTS

Eighty-six (86) male and female participants 18-88 years of age were randomly assigned to either treatment (n = 45) or control groups (n = 41).

INTERVENTIONS

Several volunteers committed to daily prayer for participants in the intervention group. Intercessory prayer commenced for 1 month and were directed toward a life concern or problem disclosed by the participant at baseline. Participants were unaware of being prayed for.

OUTCOMES MEASURES

Degree to which their problem had been resolved and the current level of concern they had about a specific life problem they described at baseline. Four component scores from the Medical Outcomes Study SF-20 were also used.

RESULTS

No direct intervention effect on the primary outcomes was found. A marginally significant reduction in the amount of pain was observed in the intervention group compared to controls. The amount of concern for baseline problems at follow-up was significantly lower in the intervention group when stratified by subject's baseline degree of belief that their problem could be resolved. Prayer intervention appeared to effectively reduce the subject's level of concern only if the subject initially believed that the problem could be resolved. Those in the intervention group who did not believe in a possible resolution to their problem did not differ from controls. Better physical functioning was observed in the intervention group among those with a higher belief in prayer and surprisingly, better mental health scores were observed in the control group with lower belief in prayer scores.

CONCLUSIONS

The results of the current study underscore the role of interpersonal belief in prayer efficacy and are consistent with the literature showing the relevance of belief in health and well-being in general. The relevance of interpersonal belief factors of the participants is recommended in future investigations.

Panic-related outcomes in patients with a personal physician

BACKGROUND AND OBJECTIVES

Although having a regular source of care is associated with positive health behaviors, its effect in specific disorders is not known. This study assessed differences in outcomes of patients with panic attacks, including symptoms, health care utilization, and treatment for panic symptoms based on the presence and specialty of a self-defined personal physician.

METHODS

In this community-based study, 97 subjects with Diagnostic and Statistical Manual, Third Edition, Revised (DSM-IIIR) panic attacks were interviewed concerning total and recent health care utilization for panic symptoms, total and current treatment for panic symptoms, and panic-related outcomes such as quality of life, psychiatric symptoms, work disability, substance abuse, and perceived control.

RESULTS

Although the absence of a personal family physician was not related to outcomes, the specialty of the personal physician was associated with significant differences in recent health care utilization and treatment for panic and in panic-related outcomes. Having a mental health provider was associated with more utilization and treatment but worse outcomes. However, such differences could generally be explained by differences in preexisting psychiatric disorders and the severity of panic symptoms.

CONCLUSIONS

Compared with having a mental health provider, having a family physician was associated with fewer recent ambulatory visits, particularly to mental health settings, and less-frequent current use of prescription, over-the-counter, or illicit drugs for panic. Yet, when corrected for panic severity and prior mental disorders, there was no difference in these outcomes.

When plagiarism becomes research

There are three possible levels of analysis in clinical research. Primary analysis deals with the original analysis of research study data. Secondary analysis is a reanalysis of the original data, either to address the original question through better techniques or to address a new question using old data. Meta-analysis is a statistical analysis of many studies done to summarize a body of literature. Meta-analysis is particularly helpful in an area in which original research studies have produced conflicting results because it enables analysis of the impact of study characteristics upon the end result. The family practitioner as a consumer of research needs to become familiar with the technique of meta-analysis because it is appearing with increasing frequency in the medical literature. Still somewhat controversial, meta-analysis requires a rigorous approach to ensure its validity: this editorial is written to assist the family practitioner in an understanding of the meta-analytic technique and point out important features that need to appear in any published meta-analysis.