The correlation between depression and BMI in elderly men compared to women
Berend Broeren
s4721381
b.broeren@student.ru.nl
Simone Hendriks
s4716159
s.hendriks@student.ru.nl
Joost Kalff
s4722175
f.kalff@student.ru.nl
Project supervisor
Dr. Rose Collard
Department Radboudumc: Psychiatry
Email: Rose.Collard@radboudumc.nl
Index
Index 2
Abstract 3
Introduction 3
Patients and methods 5
Results 6
Discussion 7
References 9
Logbook 10
Abstract
It is known the eating habits of women differs from men in stressful situations. Women tend to eat more unhealthy foods when they are stressed. The term stress-related disorders (SRD) is defined on the basis of the high vulnerability of these disorders for environmental factors such as early life adversity as well as acute recent adversity either on a psychosocial (e.g. childhood abuse) or a biological (i.e. somatic comorbidity) level. SRD include depression, anxiety and addiction.
In this paper the focus will be on depression in elderly. The difference in body mass index (BMI) between elderly men and women who suffer from depression is examined to conclude whether or not there is a different association. Data is collected from the NESDO project. Therefore, this is a cross sectional study. The statistics showed an average BMI of 26.05 kg/m² in elderly men and 26.48 kg/m² in elderly women. Subsequently the p-value was 0.331, therefore there is no significant difference of BMI in depressed elderly men and women. The null hypothesis could not be rejected.
Introduction
According to the WHO* one out of four people have been affected or will be affected by mental or neurological disorders at least once in their lives. That makes mental health problems one of the biggest health problems worldwide. Among those mental disorders depression is a severe and common disease; globally, more than 300 million people suffer from depression*. Statistics from Centraal Bureau voor Statistiek (CBS) show six percent of the elderly in the Netherlands said they suffered from depression in the past year*. This indicates depression among elderly is a severe problem. Because it is such a significant problem it is a important subject worth investigating to improve mental health.
Depression has a very strong association with stress**. Recent studies i.a.****** show that women tend to eat more unhealthy food than men in stressful situations. Most of these studies were case-controlled and showed that women eat more and more unhealthy food in stressful situations than men. For example, Grunberg N, Straub R (1992) shows that women eat 202% more sweet food when they are stressed compared to when they are not stressed. Their total caloric consumption raised with 140% in stressful situations, whereas men only ate 40% of their total caloric consumption when they are stressed compared to not stressful situations.
This study examines the effect on BMI of depression in elderly men and women. Based on previous studies, which showed that women eat more when they are stressed and the fact that depression is strongly associated with stress, one might expect a more substantial association between depression and BMI in elderly women than in men. This study clarifies whether or not there is a difference in this association in elderly men and women. Due to this clarification, this study helps to personalize the healthcare of elderly with depression.
Personalized care is an upcoming trend which is important for the improvement of healthcare. This kind of healthcare gives patients a more fitting way to cope with their disease and gives them a more accurate idea of what to expect during times of sickness. To execute personalized healthcare the difference in the effects of certain health problems between men and women, youngsters and elderly, Westerners, Asians, Africans and many more other groups must be known. Only then, one can perform proper treatments which are the most effective for every individual. This paper contributes to this goal by clarifying the difference between the association of BMI and depression in elderly men and women, hence the research question: is there a different correlation between depression and BMI in elderly men compared to women.
WHO | Mental disorders affect one in four people [Internet]. Who.int. 2018 [cited 15 June 2018]. Available from: http://www.who.int/whr/2001/media_centre/press_release/en/
Depression [Internet]. World Health Organization. 2018 [cited 8 June 2018]. Available from: http://www.who.int/news-room/fact-sheets/detail/depression
Meer dan 1 miljoen Nederlanders had depressie [Internet]. Cbs.nl. 2018 [cited 15 June 2018]. Available from: https://www.cbs.nl/nl-nl/nieuws/2016/04/meer-dan-1-miljoen-nederlanders-had-depressie
Depression [Internet]. World Health Organization. 2018 [cited 8 June 2018]. Available from: http://www.who.int/news-room/fact-sheets/detail/depression
I. BRILMAN E, ORMEL J. Life events, difficulties and onset of depressive episodes in later life. Psychological Medicine. 2001;31(05).
Zellner D, Saito S, Gonzalez J. The effect of stress on men's food selection. Appetite. 2007;49(3):696-699.
Pankevich D, Bale T. Stress and Sex Influences on Food-seeking Behaviors. Obesity. 2008;16(7):1539-1544.
Grunberg N, Straub R. The role of gender and taste class in the effects of stress on eating. Health Psychology. 1992;11(2):97-100.
Klatzkin, R., Gaffney, S., Cyrus, K., Bigus, E. and Brownley, K. (2016). Stress-induced eating in women with binge-eating disorder and obesity. Biological Psychology.
Greeno, Catherine G., Wing, Rena R. (1994). stress induced eating. psychological bulletin.
Schreiber D, Dautovich N. Depressive symptoms and weight in midlife women. Menopause. 2017;24(10):1190-1199.
Methods
The data from the data collection from the Nederlandse Studie naar Depressie bij Ouderen (NESDO) was used, which is a project that studies depressions of elderly (aged sixty and over) in the Netherlands. This is a cross-sectional study because it is an observational study which analyses data from a subset, which is representative for the population.
Concerning the data for from the NESDO project; The focus of this study was on the BMI of our patients. In the NESDO project body composition assessment included objective, standardized assessments of height, weight and hip and abdominal circumference. Based on body weight and height, BMI was calculated.
Sample
Depressed older persons were recruited from five regions: Amsterdam, Leiden, Apeldoorn/Zutphen, Nijmegen and Groningen. The recruitment took place from general practitioners and from mental health care institutes, this way persons with late-life depression in different developmental and severity stages were included. The exclusion criteria were an insufficient command of the Dutch language, dementia, suspected for dementia, or a Mini-Mental State Examination-score (MMSE) under 18/30.
The first measurement from which data was collected started in 2007 and ended in 2010. From clinics of the regional facilities for mental health, older patients who suffered from depression were recruited. The total number of patients was 326. From Amsterdam there were 113 patients recruited, from Leiden 70 patients, from Apeldoorn/Zutphen 50 patients, from Nijmegen 47 patients and from Groningen 46 patients.
All patients who were able to give written informed consent received verbal and written information about the NESDO project. Telephonic, 99% was contacted. The other 1% could not be reached by phone. In the end, 113 participated in the NESDO project. Of the people contacted by phone, 57 persons were unable to participate, because they didn’t pass the exclusion criteria. 23 persons didn’t suffer from depressive symptoms anymore, 14 persons were too sick or were physically or cognitively to frail to undergo the interview and 6 persons didn’t command the Dutch language sufficient enough. The final sample was 113, because of the 234 persons that remained, 121 refused to participate. The two groups; the one with the participating patients and the one with the patients who refused are comparable in gender, but the participating group is younger (the participating group has an average age of 72.0 years and the patients who refused had an average age of 74.2 years, p<0.03).
From 14 general practices in the regions of Amsterdam, Groningen, Leiden primary care patients were enrolled (n=52). Based on depressive complaints (n=17) or based on a positive score (4 or higher) on the fifteen-item version of the Geriatric Depression Scale (GDS-15; n=35), general practitioners referred depressed persons. 242 persons were contacted for an interview over the phone. Of those 242 persons, 58 passed the exclusion criteria and fulfilled the criteria for depression. 35 of those patients underwent in the baseline assessment because 23 refused to participate.*
The independent sample T-test was performed on the data from the NESDO project. This test is most conforming to continuous variables and shows whether or not the result is significant. The results are presented in tables.
*Comijs H, van Marwijk H, van der Mast R, Naarding P, Oude Voshaar R, Beekman A et al. The Netherlands study of depression in older persons (NESDO); a prospective cohort study. BMC Research Notes. 2011;4(1):524.
Results
Table 1. descriptive statistics
Total (n=377)
Men (n=128)
Women (n=249)
BMI, mean
26.33
26.05
26.48
BMI, range
16.46-46.16
17.57-37.98
16.46-46.16
BMI, median
25.69
25.60
25.95
BMI: Body Mass Index in kg/m2.
The baseline sample consists of 377 depressed elderly, of which 128 are men and 249 are women. The overall sample has a mean age of 70.6 years (SD: 7.3; range = 60-93). The majority (99,4 %) has a Dutch nationality. *
The mean BMI was calculated for each group and further analyzed by an independent sample T-test.
As shown in table 1, the mean BMI of men was 26.05 kg/m² with a standard deviation of 3.53 in women the mean was 26.48 kg/m2 with a standard deviation of 4.82. The mean BMI of the whole group was 26,33 kg/m2. The range and median of both groups are included in the table.
Table 2. Independent T-test
T-test for equality of means
t
df
P (2-tailed)
Mean difference
Body Mass Index (BMI)
-0.973
331.304
0.331
-0.42489
Df: degrees of freedom. Sig: significant.
The results of the independent T-test are shown in table 2. The P-value is 0.331 and the difference between the means is -0.42. The mean difference corresponds with the computed test statistics (t=-0.97). The t-value gives an indication how much the means of the two groups differ compared to how much each mean differs within the groups. Thus, a small t-value indicates a small difference in means. Because our sample size is 377 , our degrees of freedom (df) is 331.304. A higher df increases the likelihood of a significant test.
*Comijs H, van Marwijk H, van der Mast R, Naarding P, Oude Voshaar R, Beekman A et al. The Netherlands study of depression in older persons (NESDO); a prospective cohort study. BMC Research Notes. 2011;4(1):524.
Discussion
The results suggest no significant difference in the association between depression and BMI in elderly men and women.
There is a distribution of the BMI in categories. A BMI lower than 20 kg/m2 is classified as underweight, a BMI between 20 kg/m2 and 25 kg/m2 is normal, between 25 kg/m2 and 30 kg/m2 is overweight and above 30 kg/m2 is obese. When a BMI is higher than 35 kg/m2 it is called morbidly obese. The sample group has a mean BMI of 26.33 kg/m2, which means the mean of the group is classified as overweight*.
As stated in the introduction, the hypothesis was that women would have a higher BMI because they are known to eat more in stressful situations and depression is part of stress-related disorders. The results however show a different outcome (p>0.05). Therefore the null hypothesis could not be rejected.
There are studies that also studied the association of depression and obesity, however the outcomes are diversified, none of these studies had elderly as the target population.
Goldstein et al. (2016) obtained different outcomes compared to the results in this paper. This study outcome is based on overlapping brain regions that regulate mood, hypothalamic-pituitary-adrenal and -gonadal (HPA-HPG) functions. Because these areas are dense with receptors for ghrelin and leptin, which are hormones responsible for the satisfaction and hunger feelings, it could explain the comorbidity of obesity with major depressive disorders. In combination with sex-dependent effects on HPA-HPG axes, women may have a stronger association between obesity and depression.* Although this is a review and not a cross-sectional study, it does suggest a possible theory as to why women may have a stronger association between a higher BMI and depression. This paper however, does not support that outcome and is thereby in contradiction with this review.
Haukkala A, Uutela A. (1998) had similar results as this study. With a sample of more than 3,000 participants, they examined how education and gender decrease the association of obesity with cynical hostility and depression. Their outcome stated there was no significant gender difference in the associations between psychological factors and measures of obesity. * This is also a cross-sectional study and is therefore more easy to compare to this study. The results were similar to our results, which affirms our conclusion. Their sample was larger, but the most important difference is target population. Their study target population was Finnish men and women between the ages of 24 and 64 years old, while this study target population were Dutch men and women aged from 60-93.
Pratt LA, Brody DJ. (2014) found that adults with depression were more likely to be obese than adults without depression. In addition, among all persons with depression, women were more likely to be obese than men. Their research also states the age of women with depression did not influence their BMI, but for men with depression, those aged 60 and over had a higher BMI than younger men with depression.* This suggests age has a higher influence on the BMI of depressed men and may form a confounder for our study. An interesting follow-up study would therefore be to examine the influence of age on the BMI of depressed men compared to the BMI of depressed women. These studies however do differ on the fact that we did not found a correlation between a higher BMI in depressed persons compared to non-depressed persons.
In short, different studies vary in their outcomes; there are studies were no gender difference in weight is found and studies were depressed women seem to have a higher BMI than depressed men.
*Obesity and overweight [Internet]. World Health Organization. 2018 [cited 13 June 2018]. Available from: http://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight
*Goldstein JM, Holsen L, Huang G, et al. Prenatal stress-immune programming of sex differences in comorbidity of depression and obesity/metabolic syndrome. Dialogues in Clinical Neuroscience. 2016;18(4):425-436.
*Haukkala A, Uutela A. Cynical hostility, depression, and obesity: The moderating role of education and gender. International Journal of Eating Disorders. 2000;27(1):106-109.
Geeft dezelfde uitkomst als ons onderzoek. Gelijk bij man en vrouw
*Pratt LA, Brody DJ. Depression and obesity in the U.S. adult household population, 2005–2010. NCHS data brief, no 167. National Centre For Health Statistics 2014.
Confounders were not taken into account. Previous studies have shown that there are a few possible confounders that could influence our study. Age, culture, level of education, diabetes, physical limitations or use of medication such as antidepressants are the most important ones. For example, older people tend to be heavier than younger individuals. Also, generally, if your level of education is lower in comparison to someone else, the person with the lower level of education will have a higher BMI. Furthermore, if you have physical limitations it will be harder to use all the nutrients you have eaten throughout the day. *******
To draw a correct conclusion without bias, these confounders must be taken into account in a follow-up study. In addition, only the database from NESDO was used; if somehow a selection bias took place recruiting these patients, this study will also be affected by this bias.
This study was only done with elderly (>60 years). To expand this study and gain a more accurate result all ages must be included. The number of participants in this research was high enough to detect moderate or small effects in biological measures. Although there were twice as many women as men involved in this study, the number of men was sufficient enough. Another strong part of this study is that the population used was recruited from all over the Netherlands, which reduces the chance of selection bias because it makes the sample more representative for the whole population.
*O’Dea J, Dibley M. Prevalence of obesity, overweight and thinness in Australian children and adolescents by socioeconomic status and ethnic/cultural group in 2006 and 2012. International Journal of Public Health. 2014;59(5):819-828.
*Tang W e. [Study on the prevalence of childhood overweight and underweight, and the association with family socio-economic status (SES)]. – PubMed – NCBI [Internet]. Ncbi.nlm.nih.gov. 2014 [cited 14 April 2018]. Available from: https://www.ncbi.nlm.nih.gov/pubmed/24868972
*Nakamura T, Nakamura Y, Saitoh S, Okamura T, Yanagita M, Yoshita K et al. Relationship Between Socioeconomic Status and the Prevalence of Underweight, Overweight or Obesity in a General Japanese Population: NIPPON DATA2010. Journal of Epidemiology. 2018;28(Supplement_III):S10-S16.
* Nakamura T, Nakamura Y, Saitoh S, Okamura T, Yanagita M, Yoshita K et al. Relationship Between Socioeconomic Status and the Prevalence of Underweight, Overweight or Obesity in a General Japanese Population: NIPPON DATA2010. Journal of Epidemiology. 2018;28(Supplement_III):S10-S16.
*Kinge J, Morris S. Association between obesity and prescribed medication use in England. Economics & Human Biology. 2014;15:47-55.
*Kit B, Ogden C, Flegal K. Prescription Medication Use Among Normal Weight, Overweight, and Obese Adults, United States, 2005–2008. Annals of Epidemiology. 2012;22(2):112-119.
*Himmerich H, Minkwitz J, Kirkby K. Weight Gain and Metabolic Changes During Treatment with Antipsychotics and Antidepressants. Endocrine, Metabolic & Immune Disorders-Drug Targets. 2015;15(4):252-260.
For the patient and professionals, our study results mean they don’t have to expect a difference in weight in depressed men and women. Therefore, professionals do not have to instruct elderly men differently than elderly women on this subject. This clarification helps the improvement of personalized healthcare because it saves the professional and the patient time and effort. The mean BMI of the group was classified as overweight, but follow-up studies should show whether there is in fact a significant difference in the correlation between BMI and depressed persons compared to non-depressed persons. When there is, patients and professionals can anticipate on this symptom and might adjust their treatment. For now, depressed men and women do not have to adjust differently in their treatment, because there is no significant difference in BMI.
When this study is done for different age categories, different cultures, different levels of education and other variables, one can form a complete picture of how BMI might or might not differ in men and women who suffer from depression.
References
(overige literature search:
Martin-Storey A, Crosnoe R. Trajectories of overweight and their association with adolescent depressive symptoms. Health Psychology. 2015;34(10):1004-1012.
Meisjes meer dan jongens
Deze gaat over of dikke meisjes of dikke jongens meer depressieve klachten hebben dus is net een ander onderwerp.
Assari S, Caldwell C. Gender and Ethnic Differences in the Association Between Obesity and Depression Among Black Adolescents. Journal of Racial and Ethnic Health Disparities. 2015;2(4):481-493.
Deze is niet heel bijzonder)
Logbook
Datum
Verricht werk
Gemaakte afspraken
Tijdsduur
13-12-17
Schrijven introductie
3 uur
15-12-17
Schrijven methode aan de hand van de MindSet studie
Schrijven ‘innovative elements’
3 uur
2 uur
16-12-17
Maken van de planning
Schrijven ‘impact and feasibilty’
1 uur
3 uur
19-12-17
Herzien van de planning
Check van de kopjes ‘innovative elements’ en ‘impact and feasibility’
0,5 uur
3 uur
21-12-17
Maken van de samenvatting nu alle onderdelen compleet zijn
Werken aan de bronvermelding en de layout
2,5 uur
2,5 uur
22-12-17
Laatste verrichting aan eindproduct proposal (oa layout)
Inleveren project proposal
4 uur
5 minuten
15-04-18
Lijst met mogelijke confounders opgesteld
1 uur
17-04-18
Literature search naar mogelijke confounders
4 uur
24-04-18
Verbetering mogelijke confounders
2 uur
9-05-18
Afspraak met Rose Collard
1 uur
16-05-18
SPSS bestand gekregen van Rose Collard
Afspraak gemaakt om variabelen uit het bestand te zetten
1 uur
22-05-18
Lezen SPSS handleiding Alphons de Vocht
Bewerken van SPSS bestand
Daarna bewerken van het bestand
2 uur
4 uur
23-05-18
Bewerken SPSS bestand.
Bestand over opbouw van verslag bestudeerd en samengevat
2,5 uur
1 uur
29-05-18
Begonnen aan abstract
1 uur
3-06-18
begin gemaakt aan introductie
2 uur
4-06-18
afspraak Rose Collard
1 uur
8-06-18
Samen met Rose Collard de data geanalyseerd
Abstract aangepast
Afgesproken om na het weekend de eerste discussie te sturen voor feedback
1,5 uur
0,5 uur
9-06-18
De methode schrijven met behulp van het NESDO project
2,5 uur
10-06-18
Literature search
Begin gemaakt aan discussie
5 uur
3 uur
11-06-18
Samen de discussie besproken.
Abstract verbeterd
Resultaten afgemaakt.
Opgestuurd naar Rose
2 uur
0,5 uur
1,5
12-06-18
Methode bijgewerkt
Poster gemaakt
Feedback gegeven
1,5 uur
1 uur
0,5 uur
13-06-18
Paper verbeterd aan de hand van feedback van Rose Collard
De discussie en de introductie moeten nog iets worden uitgebreid
3 uur
14-06-18
Discussie aangepast
1 uur
15-06-18
Discussie aangepast
Tabel 2 beschreven in woorden
Spellingcheck
0,5 uur
0,5 uur
1 uur