Heart Mind

ORIGINAL ARTICLE
Year
: 2021  |  Volume : 5  |  Issue : 4  |  Page : 103--111

A comparative epidemiology model for understanding mental morbidity and planning health system response to the COVID-19 pandemic


David Cawthorpe 
 Alberta Health Services, Child and Adolescent Addiction and Mental Health and Psychiatry Program; Department of Psychiatry; Department of Community Health Sciences, The University of Calgary, Cumming School of Medicine, Calgary, Canada

Correspondence Address:
Prof. David Cawthorpe
Alberta Health Services, Child and Adolescent Addiction and Mental Health and Psychiatry Program, Calgary; Department of Psychiatry and Community Health Sciences, The University of Calgary, Cumming School of Medicine, Calgary; Department of Community Health Sciences, The University of Calgary, Cumming School of Medicine, Calgary
Canada

Abstract

Introduction: This particular coronavirus disease is a pandemic giving rise to great global affliction and uncertainty, even among those who have dedicated their lives to health care or the study of disease, or both. Notwithstanding those directly affected, the lives of all people have been turned upside down. Each person has to cope with her or his personal situation and a story is taking shape for everyone on earth. Coronavirus disease (COVID-19) is caused by the severe acute respiratory syndrome coronavirus 2 virus, the source of the 2020 pandemic. This paper contains brief highlights from a duplicable PubMed search of the COVID-19 literature published from January 1 to March 31, 2020, as well as a duplicable search of past influenza-related publications. Excerpts from select papers are highlighted. The main focus of this paper is a descriptive analysis of influenza and other respiratory viruses based on a 16-year population-based dataset. In addition, the paper includes analyses based on the presence or absence of mental disorder (MD) in relation to influenza and all other respiratory viruses. Methods: The investigation is descriptive and exploratory in nature. Employing a case-comparison design, a 16-year population-based dataset was analyzed to both understand the present and plan for the future. While not all viral infections are equal, this paper focuses on system responses by describing the epidemiology of respiratory viruses, such as influenza. Influenza is established in the global population and has caused epidemics in the past. Where possible direct comparisons are made between COVID-19, influenza, and other respiratory viruses. Results: Those with MD had a higher rate of viral infection per 100,000 capita compared to those with the viral infection and no MD. Further, the postviral infection MD rate was not higher compared to the MD per capita rate before viral infection. The postinfluenza rate of MD among those who were without mental disorder before influenza represents an estimate of postinfection mental health burden. Conclusions: In summary, those with preinfluenza MD are at greater risk for viral infection. Further, while the postviral infection MD rate was not higher compared to the MD per capita rate before viral infection, this independent estimate may inform the degree to which services may need to undergo a sustained increase to address the bio psychosocial needs of each served population were COVID-19 to persist and become established in the global population.



How to cite this article:
Cawthorpe D. A comparative epidemiology model for understanding mental morbidity and planning health system response to the COVID-19 pandemic.Heart Mind 2021;5:103-111


How to cite this URL:
Cawthorpe D. A comparative epidemiology model for understanding mental morbidity and planning health system response to the COVID-19 pandemic. Heart Mind [serial online] 2021 [cited 2022 Jul 3 ];5:103-111
Available from: http://www.heartmindjournal.org/text.asp?2021/5/4/103/331569


Full Text

 Introduction



Coronavirus disease (COVID-19) is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus. The first purpose of this paper is to orient mental health and psychiatric services to the current COVID-19 literature, as well as the published literature related to Influenza. The second purpose of this paper is to examine COVID-19 in comparison to the rate of influenza and all other respiratory viral (AORV) diseases on their own and in relationship to mental disorder (MD) employing a local 16-year population-based dataset.

Literature overview

A recent paper about influenza reports the following:

“One hundred years after the devastating 1918 Influenza pandemic, we still are not able to optimally prevent and treat Influenza.”[1]

However, there has been a long tradition of study into associated disease:

“For every 1000 female deaths from Influenza there was a 17% increase in births of mentally handicapped individuals 6 months later. Maternal exposure to Influenza at approximately the 3rd to 4th month of gestation may be risk factor for developing mental handicap.”[2]

At the present time there have been numerous papers about COVID-19 and some papers about mental health and disorder related to COVID-19 from which the following is an example:

“During the COVID-19 outbreak, healthcare workers and younger people were at an especially high-risk of displaying psychological impact when they spent too much time thinking about the outbreak. Continuous monitoring of the psychological consequences for high-risk populations should become routine as part of targeted interventions during times of crisis.”[3]

A reproducible PubMed literature search of the topics examined in this article are shown in [Table 1]. This review is not meant to be a compressive search in service of systematic review or meta-analysis, rather it is to illustrate the limitations of our understanding of the current pandemic and present a background against which the present contribution may, in part, answer the question: Can past population-based information about Influenza and Other Respiratory Viral infections inform our understanding and response to COVID-19?{Table 1}

In respect to search # 3 in [Table 1], the following provide examples of COVID-19 responses in respect to mental health services:

“Several changes were implemented in our acute mental health care service using a collaborative approach to maintain a balance between preventive measures to “flatten the curve” and to provide care to those who were in need.”[4]

“This study suggests that increased attention should be paid to the knowledge and attitudes of medical staff at psychiatric hospitals during the COVID-19 outbreak.”[5]

“Once mental health care institutions have developed the capabilities of serving their patients via videoconferencing and other digital technologies, there is little reason for them to give these up, in view of the many advantages…We urge practitioners to promptly start adopting e-mental health care applications, both as methods to continue their care to current patients in need and as interventions to cope with the imminent upsurge in mental health symptoms due to the coronavirus.”[6]

Of note is that some of the more recent “COVID-19 and mental issue” publications [[Table 1]: Search #3] were letters, often lacking abstracts or details beyond a title, authorship, and opinion. Many of the “influenza and mental issue papers” [[Table 1]: Search #4] were related to putative vaccination effects or willingness to vaccinate. On the upside, many publishers are presently providing free access to COVID-19-related papers.

Background

Other than the best available documentation of the rapidly expanding COVID-19 pandemic (worldometers.info) there is presently little or more than preliminary and largely speculative information related to COVID-19 in terms of baseline infection, transmission, or mortality rate. The reasons that underlay the active debate on these issues is a lack of valid and reliable information about accurate numerators and denominators employed in the reported calculations in the middle of an evolving global situation. More is the case for COVID-19 and mental issues in respect to the temporal effect of this viral infection on mental health and health service utilization. However, it is possible to establish a more informed understanding by examining population-based, system-level capacity in relation to influenza and other respiratory viral infections to compare to the per capita rates of COVID-19. In addition, this approach may take into account the effect of the temporal presence of MD. When taking into account the association of MD, it is most relevant also to examine the associated biomedical and biophysical morbidity in relation to influenza and respiratory virus infection. Evidence from the dataset employed to generate this paper has provided important information related to MD regarding putative biochemical mechanisms underpinning cancer and ulcerative colitis,[7],[8] as well as the association of MD and preventable disease.[9] In further support of reliability and validity of this methodological approach to the analysis of temporal hyper-morbidity has been the recent formation of a morbidity section in the annual World Psychiatric Association Congress.[10]

Therefore, this paper on health services utilization will report on the annual and cumulative 16-year rates of physician-diagnosed inpatient and emergency (Inpat) admissions for respiratory viral infection and influenza and the associated biomedical and biophysical morbidity. In addition, this paper presents the results of the relationship of the presence and absence of MD and diagnosed influenza or AORV infection, or both, including the respective associated biomedical and biophysical morbidity. Note that the investigation is descriptive and exploratory in nature. No formal model was proposed in advance.

 Methods



The anonymous data employed in this analysis was compiled from the public domain and from the Alberta Health Services Calgary zone repository in Alberta under the current local ethics ID: REB15-1057. All physicians in the region of study must directly bill the provincial health plan for each patient visit for payment (physician billing–[PB]). Data analyzed consisted for each billing record included an encrypted unique patient identifier, an International Classification of Diseases (ICD-9) diagnosis, a visit cost, age, and sex. The PB data were comprised of records of all individuals' visits to physicians in the region of study who sought health care for a specified problem on a specified date and subsequently were assigned an ICD diagnosis during the period of Spring 1993–Fall 2010. The dataset also integrated similar anonymous data gathered from inpatient and emergency admissions (Inpat). More descriptive details on the 16-year data source are available.[9],[11] Further details of the sample are provided in [Appendix 1].

In addition to the publicly published provincial, national, and international per capita rates of COVID-19 (worldometers. info), the historical data consisting of those with and without diagnoses of Influenza (ICD-9: Code 487–488), all other respiratory disorders (AORV; ICD-9 Code 480–486), and MDs (MD; ICD-9 code 290–319) were grouped on the following basis with the positive sign meaning “present” and the negative sign meaning “absent:”

1-Influenza-AORV-MD2-Influenza+AORV-MD3-Influenza-AORV+MD4-Influenza+AORV+MD5+Influenza-AORV-MD6+Influenza+AORV-MD7+Influenza-AORV+MD8+Influenza+AORV+MD.

Analysis

Descriptive analysis consisted of descriptive statistics represented as frequencies, means, and proportions. Graphics (bar charts, connected line graphs, and box plots) are also presented showing the annual and cumulative influenza and AORV rates and day-based COVID-19 prevalence rates represented as unique individual counts per 100,000 capita. Note that nonoverlapping whiskers between box plots indicate significant differences. For ease of visual comparison, the annual prevalence of each group was represented in some graphs as joined line plots even though each year was calculated independently and individuals might or might not have been represented in more than 1 year. Further, where viruses were present or absent or presented with or without MD, for comparison across the eight categories, where noted to be associated and for MD, the tallies of unique individuals within each group included the counts of all category-linked disorders (hyper-morbidity) within each category and the given time frames (e.g., annual vs. cumulative). Where noted as AORV or Influenza or both alone indicated only those diagnoses in singular (unlinked) counts. To compare within each graph, the box and whisker plots include the approximation of 95% confidence intervals (1.5 times the interquartile range). In comparing each category within any given graphic nonoverlapping whiskers between to box plots indicates independence (low correlation) and overlapping whiskers between to box plots indicates dependence (high correlation) to the extent of overlapping.

Similar analyses have been developed and employed in the following peer-reviewed papers.[2],[8],[9],[11],[12],[13],[14],[15]

 Results



[Figure 1] presents for a 3-week period the increase in the rate of COVID-19 per 100,000 capita in the province of study, the nation, and the world, calculated on 5 days between April 5 and 28, 2020. The provincial rate was lower, paralleled the national rate, and both were lower than the world rate.{Figure 1}

[Figure 2] shows the 16-year rates per 100,000 capita hyper-morbidity associated with AORV and Influenza in addition to the per 100,000 capita rates of AORV and Influenza diagnoses alone and the per 100,000 capita rates for the group where AORV and Influenza were diagnosed in the same individuals, but not necessarily at the same time. The box plots present the median (horizontal line roughly midway in each plot), a colored square representing the interquartile range and the whiskers representing approximately the remaining ~20% of the distribution. The position of the median and length of the whiskers indicate the skewness of the distributions.{Figure 2}

The main point of this graphic, like [Figure 3], is to illustrate the level of morbidity associated with AORV and influenza infections. The main burden of influenza and AORV is in respect to the associated biomedical and biophysical disorders. [Figure 3] shows the annual consistency of the overall 16-year hyper-morbidity associations of the per 100,000 capita rates of AORV and Influenza and where AORV and Influenza arise in the same individuals, as well as the per 100,000 capita rates of AORV and Influenza alone.{Figure 3}

While there was a downward trending across the bars representing annual AORV and influenza disease rates and their associated hyper-morbidity and the group with both influenza and AORV, the rates with AORV and influenza alone evidence greater annual variability which was more pronounced in Influenza with peaks in 1995, 1999, and 2009. H1N1 was not represented in this dataset.

[Figure 4] shows the comparison of the index MD rates per 100,000 capita before and after any index AORV or influenza diagnosis by year. Note that there was an overall downward trend in each group across the 16-year period. The rates of index MD were substantially greater before index AORV and index Influenza and lower after index AORV and index Influenza. Note that these were the conditions observed within each year where there was an index AORV or index influenza and index MD dates for individuals.{Figure 4}

[Figure 5] shows the cumulative 16-year distributions of rates per 100,000 capita temporal hyper-morbidity of MD associated with AORV and influenza. The cumulative median rates were comparable and highest when index Influenza or AORV diagnoses after any index MD over the 16-year period indicating the vulnerability of this group. Even though the cumulative 16-year rates per 100,000 capita were lower in the group where index influenza or AORV diagnoses precede any index MD, the indication being that at least some of these MDs may have resulted from Influenza or AORV infection, whether the effect was metabolic, psychosocial or a combination of both (e.g., bio-psychosocial).{Figure 5}

[Figure 6] indicates that the highest rate per 100,000 capita was for the group with MD. This was followed by the annual per 100,000 capita rate of physician diagnoses (PB) for the general population without MD or viral infection, next to the rate of MD associated with AORV, then MD associated with both AORV and Influenza, then the annual rate per 100,000 capita Influenza and AORV diagnosed in the same individuals over the course of each year, followed by AORV alone and influenza alone.

Similar to the rates for PB diagnoses in [Figure 6], in [Figure 7] the order of the groups was consistent for emergency and inpatient admissions for each group. The highest rates of emergency and inpatient admissions were associated with MD. From 1994–1999 the per 100,000 capita rates of admission were similar for those with AORV plus MD and admissions in the general population, with both slightly higher than the admission rates associated with influenza plus MD. From 2000 to 2009, the per 100,000 capita rates of admission for those with influenza or AORV, or both alone exhibited a slight downward trend.{Figure 6}{Figure 7}

The cumulative rates of the groups per 100,000 of physician diagnoses (PB) across the 16-years are illustrated in [Figure 8]. The rank order of the groups was first MD having the highest rate, followed by the rate of diagnosis in the general population, and then in MD association with each of the two viruses. These groups were followed by AORV and Influenza both together in combination with MD, then AORV and Influenza rates alone.{Figure 8}

The 16-year cumulative distributions of each group are illustrated in [Figure 9]. The rank order of the groups' distributions was the same as that shown in [Figure 8]. MD was associated with the highest emergency and inpatient cumulative admission rates, followed by the rate of emergency and inpatient admission in the general population, and then the cumulative rate of MD in association both viruses together, then each of the two viruses alone.{Figure 9}

[Figure 10] shows that the provincial and national rates remained well below the annual rates of AORV and Influenza for PB rates with and without any MD. Note that, the COVID-19 rates have a different time base, being days rather than years.{Figure 10}

[Figure 11] presents the distributions of the provincial and national (green arrow) rates per 100,000 capita rates of PB that remained well below the cumulative 16-year rates of AORV and Influenza with and without any MD. Note that the COVID-19 rates have a different time base, being days rather than years.{Figure 11}

As may be seen in [Figure 12], the COVID-19 provincial and national rates (green arrow) per 100,000 capita were lower in comparison to the emergency and inpatient rates of AORV and influenza presenting with MD and similar to, if not surpassing, the emergency and inpatient rates of AORV and influenza rates of AORV and influenza without MD. Further, the actual COVID-19 population rates of transmission, infection, identification, death, and recovery remain elusive and their veracity debated. Note that the COVID-19 rates have a different time base, being days rather than years.{Figure 12}

[Figure 13] shows the cumulative distributions of the group rates presented in [Figure 12]. The COVID-19 pandemic is in its early stages and has not yet become established in the population on an annual basis, hence the provincial and national rates per 100,000 capita of COVID-19 were low in comparison to the per 100,000 capita PB rates of AORV and influenza presenting with or without MD [Figure 10] and [Figure 11] and are only beginning to approximate the emergency and inpatient rates of AORV and influenza alone [Figure 12] and [Figure 13]. The cumulative provincial and national per 100,000 capita rates of COVID-19 to date approximated the annual rates of AORV and influenza admissions to emergency and inpatient services.{Figure 13}

[Figure 14] indicates that the provincial, national, and world cumulative COVID-19 cases have not yet surpassed the regional per 100,000 capita rates of AORV and influenza rates for unique individuals of PB and emergency and inpatient diagnoses when averaged over 16 years for April. Note that for the calculation of the regional cumulative 16-year mean values for April in this graphic, MD was not parsed into a separate category to permit direct comparison with provincial, national, and world COVID-19 rates per 100,000 capita.{Figure 14}

 Discussion



In summary, those with MD have higher rates of viral infection per 100,000 capita compared to those without viral infection. Further, while postviral infection MD rates were not higher compared to MD rates before viral infection, this estimate may inform the degree to which services may need to undergo a sustained increase, were COVID-19 to persist and become established in the population, as has influenza.

COVID-19 is a novel disease and it will take time for accurate information about case prevalence and outcomes to emerge. Its growth rate has been rapid and the upper limits of the pandemic may not necessarily be predicted at present. Nations are preparing their short-, medium-, and long-term responses to COVID-19 based on a wide variety of information and expert opinions. Notwithstanding, the historical diagnostic influenza and all other respiratory virus data may support service planning and response to the apparent hyper-morbidity, comprised of both mental and biomedical/biophysical diseases and disorders that possibly play a role in both infection vulnerability and its sequelae.

If the world history of influenza in review[1] is any indication, there may not be a COVID-19 cure anytime in the near future. Hence, antiseptic and hygiene practices, along with personal protection devices in some instances, are the principal means of prevention, together with limiting contagion via maintaining adequate personal space, and identification of infected individuals and vulnerable groups, with, as required, social control, quarantine of the exposed, and appropriate medical support of the infected. Such prompt practices identified in some countries, such as Taiwan, have been both timely and exemplar in curbing the spread of COVID-19.

A study of the record for AORV and Influenza infection provides a basis of comparison for examining the emergence of COVID-19 and, perhaps, more important for service planning, insight into the extensive degree of associated hyper-morbidity, or in other words, system-burden. There is little if any systematic population-based study of the prodromal vulnerability to or sequelae of respiratory viral infections, notwithstanding that the highest death rates are among those with multi-morbidity.

Examination of the temporal hyper-morbidity of MD in association with AORV or influenza infection found the rates of MD to be higher preinfection, whereas the postinfection rates provide a measure of the degree to which system-response may be predicted and planned. The proportion of MD postviral infection represents an increase in time and, while lower than the overall rate of previral MD, is to some degree additive. The additional postviral MD cases and may be used to help mental health services anticipate and model the required up-regulation of services required to meet the need of the affected community. For example, those with MD before influenza or AORV, or both, may require more service, even after the onset and resolution of the viral infections. In the context of the degree of hyper-morbidity across the eight groups, MD is a linchpin on which the greatest system-level burden turns.

Further, the methods underpinning the findings of this study are generalizable in terms of system-level planning. The example of influenza and AORV with and without MD may help services plan to meet the long-term needs of those who do not necessarily have MD, but are nevertheless at risk for unanticipated disorders, such as COVID-19. The example indicates the value of the profile of temporal hyper-morbidity for the range of diseases and disorders associated with AORV and Influenza infection, and curently COVID-19.

The profile of prodromal diseases and disorders arising before viral infection may inform services about vulnerability. The disease profile of sequelae may inform health services planning with precision. For example, population-based study of temporal hyper-morbidity focusing on autism has illustrated this utility.[11] Further, two studies focusing on the relationship of the temporal hyper-morbidity association of MD with cancer and ulcerative colitis have pointed to putative mechanisms giving rise to vulnerability.[7],[8]

The algorithms underpinning this analysis may be operationalized in near-real-time, permitting any division of medicine to measure, interpret, understand, and plan regional and provincial responses to the effects of the current COVID-19 pandemic on the population in general, or in relation to vulnerable groups consisting of any complex, class, or range of diseases or disorders.

The SARS-CoV-2 virus which causes the COVOID-19 disease has been estimated to mutate about every 4 days.[16] Similar to the common cold, another coronavirus, there may not be a rapid solution to a cure, unlike vaccination against childhood diseases. It may also be unlikely that the world will completely return to the way it was even over a considerable period. Hence, it behooves mental health services and other divisions of health and medicine to prepare for and adapt to the present shape of the world; a shape that the world presently demands.

The values of [Table 2] are ranked from the highest to lowest 16-year mean cost per patient. The ranking is logical in its order related to hyper-morbidity, moving from no presence of either virus or any mental disorder (lowest) through to the presence of all three disease states (highest), with AORV being less frequent and more costly in the presence of Mental Disorder. Less frequently AORV and Influenza arose in the same indivduals with Mental Disorder, but at a greater average cost{Table 2}

Limitations

Note that there were various data limitations from all sources of data. With respect to the population sample, limitations have been detailed in the peer-reviewed publications based on this dataset.[12],[13],[14] In defense of the threats to validity in the population dataset, the annual rates of MD, Influenza and AORV diagnoses have been relatively consistent of the 16 years with a slight downward trend in each. In respect to COVID-19, underestimated case identification, low levels of COVID-19 testing and accuracy of existing test kits, potential over-diagnosis, and inaccurate COVID-19-attributed mortality remain issues influencing the estimates of the “true” population values of the numerators and denominators.

In alignment with current capacity-building strategies in the regional health zone developed under the rubric of shaping demand,[17] adaptation to the current COVID-19 pandemic might involve, as recommended in the paper, the development of mental health positions within the existing budgets of all medical divisions. This strategy is salient, given observed 16-year hyper-morbidity for all diseases associated with MD and especially now those associated with viral infections. The temporal association of MD with AORV and Influenza is informative, especially with the expected increase of bio-psychosocial sequelae anticipated to become associated with the COVID-19 pandemic.

 Conclusions



Preinfluenza MD puts individuals at greater risk for viral infection. Further, while the postviral infection MD rate was not higher compared to the MD per capita rate before viral infection, this independent estimate may inform the degree to which services may need to undergo a sustained increase to address the bio-psychosocial needs of each served population was COVID-19 to persist and become established in the global population. This population-based analysis was conducted early in the COVID-19 pandemic, shortly after nations, such as Canada, began to respond to the World Health Organization's January 2020 announcement and commence containment actions in March 2020. Since that time, the world has learned about the devastating effects of slow pandemic response and re-opening to soon for economic reasons. The evidence for this in the number of waves experienced to date. Our behavior naturally facilitates the emergence of variants; viruses that can “jump” where possible will jump, infect, and reproduce. Whether or not they “jump” euphemistically refers to human behavior permitting transmission. Countries such as China and Taiwan, no strangers to national and global emergencies, are to be commended for acting rapidly and decisively following the World Health Organization's COVID-19 pandemic announcement in January 2021. Western countries have much to learn about protecting the global community.

The present paper may help those with relevant information to plan services accordingly. Examining mental problems in terms of their intimate life-span relationship to biomedical and biophysical disorders helps to illustrate an integrated system of care approach. For example, a similar more fine-grained prospective analysis employing the same dataset of vulnerability to viral pneumonia itself led to a national presentation that described an evidence-based framework for rationing limited amounts of COVID-19 vaccine.[15]

Ethic statement

The anonymous data employed in this analysis was compiled from the public domain and from the Alberta Health Services Calgary zone repository in Alberta under the current local ethics ID: REB15-1057, oringinally granted in May 2009 and renewed in August 2021.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

 Appendix 1: Data set description (Tables 3–6)



[INLINE:1]

[INLINE:2]

[INLINE:3]

[INLINE:4]

References

1Krammer F, Smith GJ, Fouchier RA, Peiris M, Kedzierska K, Doherty PC, et al. Influenza. Nat Rev Dis Primers 2018;4:3.
2Takei N, Murray G, O'Callaghan E, Sham PC, Glover G, Murray RM. Prenatal exposure to influenza epidemics and risk of mental retardation. Eur Arch Psychiatry Clin Neurosci 1995;245:255-9.
3Huang Y, Zhao N. Mental health burden for the public affected by the COVID-19 outbreak in China: Who will be the high-risk group? Psychol Health Med 2021;26:23-34.
4Kavoor AR, Chakravarthy K, John T. Remote consultations in the era of COVID-19 pandemic: Preliminary experience in a regional Australian public acute mental health care setting. Asian J Psychiatr 2020;51:102074.
5Shi Y, Wang J, Yang Y, Wang Z, Wang G, Hashimoto K, et al. Knowledge and attitudes of medical staff in Chinese psychiatric hospitals regarding COVID-19. Brain Behav Immun Health 2020;4:100064.
6Wind TR, Rijkeboer M, Andersson G, Riper H. The COVID-19 pandemic: The 'black swan' for mental health care and a turning point for e-health. Internet Interv 2020;20:100317.
7Cawthorpe D, Davidson M. Temporal comorbidity of mental disorder and ulcerative colitis. Perm J 2015;19:52-7.
8Cawthorpe D, Kerba M, Narendran A, Ghuttora H, Chartier G, Sartorius N. Temporal order of cancers and mental disorders in an adult population. BJPsych Open 2018;4:95-105.
9Chartier G, Cawthorpe D. From 'Big 4' to 'Big 5': A review and epidemiological study on the relationship between psychiatric disorders and World Health Organization preventable diseases. Curr Opin Psychiatry 2016;29:316-21.
10Heinze G, Sartorius N, Guizar Sanchez DP, Bernard-Fuentes N, Cawthorpe D, Cimino L, et al. Integration of mental health comorbidity in medical specialty programs in 20 countries. Int J Psychiatry Med 2021;56:278-93.
11Cawthorpe D. A 16-year cohort analysis of autism spectrum disorder – Associated morbidity in a pediatric population. Front Psychiatry 2018;9:635.
12Ghuttora HK, Cawthorpe D. Treatment of physical disorder in children with mental disorder: A health care utilization study. J Hosp Adm 2013;3:24-31.
13Wilkes TC, Guyn L, Li B, Lu M, Cawthorpe D. Association of child and adolescent psychiatric disorders with somatic or biomedical diagnoses: Do population-based utilization study results support the adverse childhood experiences study? Perm J 2012;16:23-6.
14Cawthorpe D. A novel population-based health index for mental disorder. Perm J 2013;17:50-4.
15Cawthorpe D. Population-based model for rationing COVID-19 vaccine: Diagnostic profiles indicating risk. In: Apple Books, Cupertino, California, USA. 2021. p. 1-16.
16Benvenuto D, Giovanetti M, Salemi M, Prosperi M, De Flora C, Alcantara LC Jr., et al. The global spread of 2019-nCoV: A molecular evolutionary analysis. Pathog Glob Health 2020;114:64-7.
17Melathopolous K, Cawthorpe D. Impact of central intake development and system change on per capita child and adolescent mental health discharges from 2002 to 2017: Implications for optimizing system design by shaping demand. Perm J 2019;23:1-24.