Quality of life of older women and men in situations of intimate partner violence

Abstract

The present paper sought to investigate the association between intimate partner violence (IPV) and the levels of quality of life (QoL) and its domains (control and autonomy; personal fulfillment and pleasure), in older people. A population-based cross-sectional study of the second wave (2013/2014) of the EpiFloripa Study was carried out with older people (n = 649) living in Florianópolis, SC, Southern Brazil. The average QoL score and their domains were described according to the adjustment variables (age, income, cognitive deficit, depressive symptoms, and dependence on activities of daily living) stratified by gender. The relation between IPV and QoL was analyzed using multiple linear regression with a statistical significance of 5%. The average scores of QoL, personal fulfillment, and pleasure were similar between genders, while control and women’s autonomy were significantly lower compared to men (p= 0.04). Lower QoL scores were observed in women exposed to violence in the three directionalities analyzed: perpetrated (-3.15; 95%CI: -4.84; -1.45), bidirectional (-2.59; 95%CI: -4.10; -1.09), and suffered (-1,62; 95%CI: -3.06; -0.17); the most affected ones were those who were aggressive. The control and autonomy were affected for the perpetrators and those involved in bidirectional violence, while lower scores of personal fulfillment and pleasure were seen among the older women in any violent situation. Men did not have their QoL impaired due to IPV, neither as victims nor perpetrators of this violence. It was concluded that IPV has an asymmetric impact on the QoL of older people when it comes to gender with women being the most affected ones.

Keywords
Intimate Partner Violence; Quality of life; Men; Women; Health of the Elderly

Resumo

Este artigo buscou investigar a associação entre a violência por parceiro íntimo (VPI) e os níveis de qualidade de vida (QV) e seus domínios (controle e autonomia; realização pessoal e prazer), em pessoas idosas. Realizou-se estudo transversal de base populacional da segunda onda (2013/2014) do Estudo EpiFloripa, com pessoas idosas (n=649), residentes em Florianópolis, SC, Sul do Brasil. As médias de QV e seus domínios foram descritas segundo as variáveis de ajuste (idade, renda, deficit cognitivo, sintomas depressivos e dependência de atividades de vida diária), estratificadas por sexo. Analisou-se a associação entre VPI e QV por meio de regressão linear múltipla, com significância estatística de 5%. Os escores médios de QV, bem como realização pessoal e prazer foram semelhantes entre os sexos, enquanto o controle e autonomia das mulheres foram significativamente inferiores em comparação aos homens (p=0,04). Observou- se menores escores de QV nas mulheres expostas à situação de violência, nas três direcionalidades analisadas: perpetrada (-3,15; IC95%: -4,84; -1,45), bidirecional (-2,59; IC95%: -4,10; -1,09) e sofrida (-1,62; IC95%: -3,06; -0,17) e as mais prejudicadas foram aquelas que cometeram agressões. O controle e autonomia foi afetado para as perpetradoras e envolvidas na violência bidirecional, enquanto menores escores de realização pessoal e prazer foram verificados entre as idosas, em quaisquer situações de violência. Os homens não tiveram sua QV prejudicada por estarem em situação de VPI, tanto vítimas como perpetradores dessa violência. Concluiu-se que a VPI possui impacto assimétrico sobre a QV de pessoas idosas em relação ao sexo, prejudicando com maior intensidade as mulheres.

Palavras-Chave:
Violência por Parceiro Íntimo; Qualidade de Vida; Homens; Mulheres; Saúde do Idoso

INTRODUCTION

With the sharp aging of the population, it is necessary to provide older people with positive experiences and quality of life during this period. Therefore, they need to be free from all forms of violence and live in safety and dignity¹. Intimate partner violence (IPV) is defined as any act of violence, whether physical, sexual, psychological, or economic abuse².

Exposure to violence can result in unnecessary suffering, injury or pain, loss or violation of human rights, post-traumatic stress disorder, and somatization, tending to severely affect the physical and mental health of the victims, which influence the quality of life (QoL) at lower levels³. For older people, QoL is defined by satisfaction in four domains: control - related to the ability to actively intervene in their environment -, autonomy - related to the right to be free from unwanted interference from others -, pleasure - which refers to the search for pleasant activities -, and lastly personal fulfillment - describing the full personal development⁴,⁵.

To reach QoL, social relationships are emphasized, including well-being in intimate relationships. However, although still disguised among older people, IPV presents as a form of communication and can be an alternative to conflict resolution⁶, which often starts in adulthood.

IPV is a human rights violation with great magnitude in the world population⁷. Studies estimate that more than 30% of women and about 25% of men suffer this type of violence worldwide⁸, although victimization of men is less investigated and explored in the literature. The analysis of physical IPV prevalence showed that more men (6.4%) suffered said aggressions in Denmark compared to women (5.0%)⁹. This finding exposes the aggressions against men, transcends the unidirectionality of violence, and reinforces the need to investigate both genders as probable victims or aggressors in the intimate relationship.

The analysis of the effect of physical⁹and psychological violence and controlling behavior in intimate partners showed a loss of QoL of older women. A Danish study exploring the association between IPV and QoL in different age groups concluded that older people had a greater QoL reduction when exposed to violence by their partners compared to adults, emphasizing the relevance of investigating the phenomenon also in this age group.

Although studies on the effects of IPV on the QoL of older people are still scarce in the literature, the subject has been gradually explored in European studies³.⁹ and North American studies¹⁰ in which the context of aging is more present. However, none of the studies analyzed the impact of IPV on QoL and its domains, according to the directionality of violence and the gender of the older people. In Brazil, no publications were found on the topic referring to the older population, which explains the gap of knowledge and the novelty of the present study. In this context, we seek to investigate the association between IPV (suffered, perpetrated, and bidirectional) and the levels of QoL in its domains (control and autonomy, personal fulfillment, and pleasure) in older women and men living in Florianópolis, Santa Catarina.

METHOD

This cross-sectional study is part of a longitudinal population and household study carried out with older people (60 years and over) living in the urban area of Florianópolis, in the State of Santa Catarina, Southern Brazil. The data in the present study come from the research ‘Condições de Saúde de Idosos de Florianópolis’ (Health Conditions of Older People in Florianópolis), the so-called EpiFloripa¹¹,¹². According to the 2010 demographic census, the municipality had a total population of 421,239 inhabitants, with older people corresponding to 11.4% of this total. The municipal human development index (MHDI) was 0.847 that same year, which is considered high and takes third place among the municipalities and the first among the Brazilian capitals¹³,¹⁴.

The baseline sample was selected in two stages by conglomerate. The first unit was the census tracts. Florianópolis has 420 urban tracts, 80 of which drawn systematically corresponding to 8 tracts in each income decile (R$192.80 - R$13,209.50). The units in the second stage were households. It was necessary to update the number in each tract (enrollments) and only residential addresses permanently occupied were registered. The number of households ranged from 61 to 725, and small tracts were grouped to reduce the variation coefficient from 52.7% (n = 80 tracts) to 35.2% (n = 83 tracts) taking into account the geographic location and the corresponding income decile, as well as the division of very large areas. It was estimated that 60 families per tract would need to be visited.

The sample size was estimated using the EpiInfo program version 6.04 (Centers for Disease Control and Prevention, Atlanta, USA). It was based on the prevalence calculation formula and the parameters of population size (44,460), confidence level (95%), unknown prevalence (50%), sample error (4 percentage points), design effect (estimated at 2), plus 20% for estimated losses and 15% for association studies. This resulted in a minimum sample of 1,599 individuals. The sample was expanded to 1,911 individuals due to the availability of resources, but 1,702 older people were interviewed. Of those respondents of the first wave, 376 losses (22.1%) were identified, which included 217 deaths and 129 refusals (7.6%), accounting for 1,197 participants in 2013/2014 (response rate of 70.3%).

Data were collected through individual interviews conducted by trained interviewers. Validated instruments were used and a pilot study was developed (n = 99 in the first wave, n = 76 in the second). The interviews were recorded on portable digital devices (PDA) in 2009/2010, and netbooks in 2013/2014.

The quality control of the EpiFloripa Idoso study was carried out by a short telephone questionnaire (with eight questions) to about 10% of the older people in the sample selected at random. The reproducibility of the questions showed satisfactory to a good agreement (first wave, kappa between 0.6 and 0.9; and second wave, kappa between 0.5 and 0.9).

For the present study, the following inclusion criteria were applied to the sample of the second wave: the older person must have answered exclusively in full the QoL (CASP-19)⁴ and IPV (CTS-1)¹⁴ questionnaires, and have had an intimate partner during the last twelve months. Of the 1,197 participants, 458 had no intimate partner during the last 12 months, 57 interviews were answered by informants, there were 31 refusals to respond to CTS-1, and 2 incomplete CASP-19 interviews, accounting for 649 eligible older people with a response rate of 54.2%. The sample flowchart is shown in Figure 1.

Figure 1
Flowchart of the study sample size. Epi Floripa Study. Florianópolis, SC, 2013/2014.

In the present study, the outcome quality of life (QoL) was measured using the CASP-19 instrument. This scale comprises four domains divided into 19 items: control (age is a limitation; there is no control over what happens to the individual; they feel free to plan their future; they feel excluded from everything), autonomy (they can do whatever they want; family responsibilities prevent them from doing what they want; they feel free to do stuff; health conditions and lack of money are limitations), personal fulfillment (they feel excited; they realize life has a meaning; they like what they do; they like the company of others; they feel happy looking at the past), and pleasure (they feel full of energy; they choose to do new things; they are satisfied with the direction that life has taken; they feel that life is full of opportunities and the future looks good). Such domains have the same level of importance, there is no hierarchical organization. For each item, there are four response options on the Likert scale (often, sometimes, rarely, never). A score was assigned to each response, the overall score of CASP-19 ranges from 0 representing the complete absence of QoL to 57 when there is full satisfaction⁴.

QoL was measured by the total score (CASP-19), and its domains were grouped into control and autonomy (scores from 0 to 27), and personal fulfillment and pleasure (scores from 0 to 30). This division has theoretical support in the literature¹⁵,¹⁶ where it was found that the four domains of CASP-19 were not sufficiently distinct for their isolated analysis. Psychometric analyzes carried out in Eastern Europe¹⁷ and Ireland¹⁸ concluded that control and autonomy related to the individual ability to initiate and achieve goals, and personal fulfillment and pleasure representing the full reach of the human potential, when grouped, better represent the QoL of the older people.

The exposure variable was the IPV measured by the cross-cultural adaptation of the instrument Conflict Tactics Scales Form R (CTS-1)¹⁵ developed to measure violence between the couple. The questionnaire investigates the presence of verbal aggression in 6 items of insults and threats (cursed or insulted, sulked, left the place, did/said things to irritate, threatened to hit or throw things, destroyed/threw objects) and physical aggression in 9 items of physical or explicit strength (throwing objects, pushing/grabbing, slapping, kicking, biting or punching, hitting or trying to hit with objects, spanking; strangling/suffocating, threatening with knife or weapon). It was possible to verify the directionality of violence as the interviewee was asked if they committed the act against their partner (perpetrated violence) and if the partner committed it against them (suffered violence). When violence was suffered and perpetrated by the same individual, it was classified as bidirectional violence. The presence of IPV was considered when the response was positive for at least one of the items on the scale in the reminiscent period for the last 12 months. This variable was transformed into a dummy variable considering values equal to 0 or 1 and stratifying the sample into yes and no. The CTS-1 was used in other Brazilian studies¹⁹,²⁰ with good reliability and a low refusal rate.

The adjustment variables were age group (60-69, 70-79 and 80 years and over), family income per capita in minimum wages (<1, 1-5, 5-10,> 10), cognitive deficit (none or probable), depressive symptoms (none or suspected depression), dependence on activities of daily living (ADL) categorized as absent, mild, moderate, severe. The cognitive deficit was measured by Folstein’s Mini-Mental State Examination (MMSE)²¹, the presence of depressive symptoms was assessed by the Geriatric Depression Scale (GDS-15)²², and dependence on ADL by the Scale of Activities of Daily Living by BOMFAQ/OARS²³.

Initially, descriptive statistics of the adjustment variables (age group, income, cognitive deficit, depressive symptoms, and dependence on ADL) were performed, presenting absolute and relative frequencies stratified by gender. The QoL level was presented by the total score and the domains comprising control and autonomy, personal fulfillment, and pleasure. Measures of central tendency (average) and dispersion (standard deviation) were used according to the adjustment variables for men and women. QoL was expressed by average and 95% confidence intervals (95%CI) according to the independent variables. The averages were compared using the Student t (gender), Mann-Whitney (depressive symptoms and cognitive deficit), and Kruskal-Wallis tests (age group, income, and dependence on ADL), and non-parametric trend for the variables of age, income, and dependence on ADL.

The IPV variables (suffered, perpetrated, and bidirectional) and the QoL scores and their grouped domains (control and autonomy, personal fulfillment and pleasure) were analyzed using multiple linear regression. In the unadjusted and adjusted analysis, results were presented in beta coefficient (β) and their respective 95% confidence intervals (95%CI). Three regression models were conducted: total QoL score, control and autonomy, and personal fulfillment and pleasure stratified by gender. In the adjusted analysis, each exposure referring to IPV (suffered, perpetrated, and bidirectional) was controlled by the adjustment variables to estimate its effect on the QoL scores and domains. The level of statistical significance was set at 5% for the association.

In the adjusted linear regression models, à the residues were analyzed by the evaluation of the heteroscedasticity and normality, the verification of the standard residues, and the variance inflation factor (VIF). For data analysis, statistical software was used. The effect of the sample design by conglomerates was considered, and the sample weights were incorporated.

The EpiFloripa Idoso study was approved by the Human Research Ethics Committee of Universidade Federal de Santa Catarina (protocol 352/2008 and CAAE 16731313.0.0000.0121). All participants signed the free and informed consent terms.

RESULTS

Of the 1,197 older people in the second wave, 649 were part of the study with a response rate of 54.2%. The study participants (n=649) had higher QoL scores (46.3) when compared to non-participants (44.9), with statistical significance (p=0.008). The majority of non-participants (n=548) were female (85.2%) older than 70 years (76.4%) with less than 4 years of education and income below five minimum wages, with a higher prevalence of depression, cognitive deficit, and dependence on ADL when compared to the participants.

The descriptive analysis of the sample and average QoL scores and their domains according to demographic, socioeconomic, and health conditions in men and women are described in Table 1.

Table 1
Descriptive analysis and quality of life scores stratified by gender according to demographic, socioeconomic, and health conditions in women and men. EpiFloripa Idoso, Florianópolis, Santa Catarina, 2013/2014.

			CASP-19		Control and Autonomy		Personal fulfillment and Pleasure
	Men	Women	Men	Women	Men	Women	Men	Women
	n (%)	n (%)	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)
	340 (52,4)	309 (47,6)	46,8 (7,5)	45.8 (8,1)	21,3 (4,6)	20,5 (5,0)	25,5 (4,2)	25,2 (4,2)
p***			0,093		0,042		0,405
Age group (years)	n= 340	n= 309
60- 69	135 (37.5)	147 (47.2)	47.9 (6.9)	46.6 (7.8)	21.9 (4.2)	20.8 (5.0)	26.0 (4.3)	25.8 (4.0)
70- 79	146 (45.1)	129 (42.0)	46.5 (7.4)	44.8 (8.4)	21.2 (4.6)	20.1 (5.3)	25.3 (4.0)	24.7 (4.3)
80 or more	59 (17.4)	33 (10.8)	44.9 (8.7)	45.6 (7.7)	20.1 (5.1)	21.1 (4.7)	24.7 (4.4)	24.4 (4.5)
p*			0.053	0.171	0.084	0.491	0.030#	0.016#
Income (minimum wage)	n= 329	n= 288
Less than 1	9 (3.3)	15 (4.2)	45.4 (7.2)	43.1 (8.6)	21.2 (3.7)	19.0 (5.4)	24.1 (4.5)	24.0 (4.0)
Between 1- 5	138 (38.1)	142 (50.5)	45.8 (8.6)	44.9 (8.5)	21.0 (5.0)	20.1 (5.4)	24.8 (4.8)	24.8 (4.6)
Between 5- 10	77 (24.8)	76 (26.3)	46.7 (6.4)	45.9 (7.3)	21.4 (4.4)	20.9 (4.2)	25.2 (3.5)	25.0 (4.0)
Greater than 10	105 (33.8)	55 (19.0)	48.0 (6.8)	48.5 (6.7)	21.5 (4.2)	21.6 (4.9)	26.4 (3.8)	26.9 (2.9)
p*			0.242	0.021#	0.933	0.166	0.003#	0.003#
Cognitive Deficit	n= 340	n= 307
No	290 (87.3)	243 (80.0)	47.4 (6.9)	46.3 (7.8)	21.6 (4.2)	20.8 (5.0)	25.7 (3.9)	25.5 (4.1)
Probable	50 (12.7)	65 (20.0)	43.3 (9.7)	43.7 (8.8)	19.3 (5.9)	19.5 (5.4)	24.0 (5.2)	24.2 (4.5)
p**			0.005	0.021	0.016	0.075	0.021	0.033
Depressive Symptoms	n= 340	n= 308
No	290 (85.7)	243 (80.8)	48.5 (5.9)	48.1 (6.7)	22.0 (4.2)	21.9 (4.3)	26.4 (3.0)	26.2 (3.5)
Suspected depression	50 (14.3)	65 (19.2)	36.6 (8.3)	37.1 (7.2)	16.7 (4.4)	15.5 (4.5)	19.8 (5.9)	21.7 (4.9)
p**			< 0.001	< 0.001	< 0.001	< 0.001	< 0.001	< 0.001
Dependence on activities of daily living	n= 340	n= 309
Absence	146 (43.2)	83 (26.7)	51.0 (4.1)	50.6 (5.5)	23.6 (3.0)	23.5 (3.8)	27.3 (2.2)	27.1 (2.6)
Mild	135 (39.6)	138 (47.1)	45.4 (7.6)	46.4 (7.4)	20.7 (4.3)	21.0 (4.3)	24.7 (4.6)	25.4 (4.4)
Moderate/Severe	59 (17.2)	88 (26.2)	39.6 (7.5)	40.2 (8.0)	16.8 (4.6)	17.1 (5.2)	22.7 (5.1)	23.1 (4.3)
p*			< 0.001#	< 0.001#	< 0.001#	< 0.001#	< 0.001#	< 0.001#

• * Kruskal-Wallis test; ** Mann-Whitney test; *** Student’s t-test, comparison of average quality of life and domains for men and women; # trend according to non-parametric trend test for the ordinal groups.

The majority of participants were men (52.4%), 45.1% of which aged 70-79 years, 58.6% with family income above five minimum wages, and 43.2% had no dependence on ADL. Among women, the age group of 60-69 years (47.2%), family income of 1 to 5 minimum wages (50.5%), and dependence on light ADL (47.1%) were predominant. For both genders, depressive symptoms and a probable cognitive deficit in most individuals were absent.

The average QoL scores measured was 46.8 (SD=7.5) among men, and 45.8 (SD=8.1) among women (p=0.093). Women had significantly lower scores (20.5; SD=5.0) for control and autonomy when compared to men (21.3; SD=4.6) (p=0.042). In the domain of personal fulfillment and pleasure, the averages were 25.5 (SD=4.2) for men and 25.2 (SD=4.2) for women (p=0.405), with no statistically significant difference between genders.

The overall QoL was higher according to the income only for women (p=0.021). The levels of personal fulfillment/pleasure were higher both for men and women according to increased income and lower according to increased age, presenting a linear trend. In the presence of depressive symptoms, cognitive deficit, and dependence on ADL, lower QoL scores and their domains were found, except for control/autonomy in women.

When measuring the average scores of QoL, women in a situation of IPV in the three directions analyzed (suffered, perpetrated, and bidirectional) showed lower levels of QoL. Women who perpetrated IPV had lower QoL scores ( -3.15; 95%CI: -4.84; -1.45), followed by those involved in bidirectional IPV (-2.59; 95%CI: -4, 10; -1.09) and those who suffered violence (-1.62; 95%CI: -3.06; -0.17) in the adjusted analysis (Table 2). nThe QoL is compromised when women are involved in a situation of IPV, especially when they are responsible for acts of aggression.

Table 2
Unadjusted and adjusted analysis of the total quality of life score (CASP-19) according to intimate partner violence suffered, perpetrated, or bidirectional stratified by gender. EpiFloripa Idoso, Florianópolis, Santa Catarina. 2013/2014.

	Unadjusted analysis				Adjusted analysis*
	Quality of Life - CASP 19
	Men		Women		Men		Women
	β (95%CI)	p	β (95%CI)	p	β (95%CI)	p	β (95%CI)	p
Suffered violence
No	1		1		1		1
Yes	0.14 (-2.16;2.45)	0.901	-2.37 (-4.37; 0.37)	0.022	0.31 (-1.24;1.86)	0.691	-1.62 (-3.06;0.17)	0.021
Perpetrated violence
No	1		1		1		1
Yes	-0.73 (2.33; 0.86)	0.364	-3.90 (-5.66; 2.15)	<0.001	-0.08 (-1.38;1.21)	0.900	-3.15 (4.84; -1.45)	<0.001
Bidirectional violence
No	1		1		1		1
Yes	0.35 (-1.90; 2.62)	0.751	-3.58 (-5.58; -1.59)	0.001	0.31 (-1.27;1.91)	0.693	-2.59 (-4.10; -1.09)	0.001

• * Model adjusted by age group, income, cognitive deficit, depressive symptoms, and dependence on activities of daily living; 95%CI = 95% confidence interval.

When analyzing the adjusted QoL domains, the control and autonomy scores were lower among women who perpetrated violence (-1.62; 95%CI: -2.70; -0.55) and for those in situations of bidirectional violence (-1.36; 95%CI: -2.41; -0.31) (Table 3). Regarding the domain of personal fulfillment and pleasure, the scores were lower for women perpetrating violence (-1.52; 95%CI: -2.50; -0.54), followed by those involved in bidirectional violence (-1.23; 95%CI: -2.12; -0.34) and those who suffered IPV (-0.81; 95%CI: -1.58; -0.04) in the adjusted analysis (Table 4).

Table 3
Unadjusted and adjusted analysis of the domain of quality of life - control and autonomy - according to intimate partner violence suffered, perpetrated, or bidirectional stratified by gender. EpiFloripa Idoso, Florianópolis, Santa Catarina, 2013/2014.

	Unadjusted analysis					Adjusted analysis*
	Control and Autonomy
	Men			Women		Men		Women
	β (95%CI)		p	β (95%CI)	p	β (95%CI)	p	β (95%CI)	p
Violence suffered
No	1			1		1		1
Yes	0.09 (-1.13;1.33)	0.870		-1.22(-2.57; 0.11)	0.073	0.20 (-0.66;1.02)	0.637	-0.80 (-1.89; 0.27)	0.140
Perpetrated violence
No	1			1		1		1
Yes	-0.32 (-1.28; 0.62)	0.490		-2.18 (-3.24;-1.11)	<0.001	-0.01 (-0.81; 0.77)	0.963	-1.62 (-2.70; -0.55)	0.003
Bidirectional violence
No	1			1		1		1
Yes	0.15 (-1.03; 1.39)	0.801		-1.95 (-3.20;-0.69)	0.002	0.07 (-0.83; 0.98)	0.872	-1.36 (-2.41; -0.31)	0.012

• * Model adjusted by age group, income, cognitive deficit, depressive symptoms, and dependence on activities of daily living; 95%CI = 95% confidence interval.

Table 4
Unadjusted and adjusted analysis of the domain of quality of life - personal fulfillment and pleasure - according to intimate partner violence suffered, perpetrated, or bidirectional stratified by gender. EpiFloripa Idoso, Florianópolis, Santa Catarina, 2013/2014.

	Unadjusted analysis				Adjusted analysis*
	Personal fulfillment and Pleasure
	Men		Women		Men		Women
	β (95%CI)	p	β (95%CI)	p	β (95%CI)	p	β (95%CI)	p
Suffered violence
No	1		1		1		1
Yes	0.02 (-1.23; 1.28)	0.964	-1.14 (-2.10; -0.18)	0.022	0.10 (-0.84; 1.05)	0.821	-0.81 (-1.58; -0.04)	0.035
Perpetrated violence
No	1		1		1		1
Yes	-0.42 (-1.34;0.50)	0.371	-1.72 (-2.70; -0.74)	0.011	-0.06 (-0.90; 0.77)	0.872	-1.52 (-2.50; -0.54)	0.003
Bidirectional violence
No	1		1		1		1
Yes	0.18 (-1.01; 1.38)	0.766	-1.63 (-2.68; -0.59)	0.002	0.24 (-0.68; 1.17)	0.697	-1.23 (-2.12; -0.34)	0.007

• * Model adjusted by age group, income, cognitive deficit, depressive symptoms, and dependence on activities of daily living; 95%CI = 95% confidence interval.

Regarding men, the unadjusted and adjusted analyzes showed that there were no differences in the QoL score and its domains among those in situations of IPV in the three directions analyzed compared to those not exposed to violence. Note that experiencing IPV brought negative effects only on the QoL of women, while men were not affected.

In the analysis of residues, it was found that they showed normal distribution, there were no specification errors in the regression models according to the analysis of standard residues, and multicollinearity was not identified using the variance inflation factor (VIF).

DISCUSSION

The present study is relevant to understand the association between IPV and QoL in older people in Brazil by the analysis of suffered, perpetrated, and bidirectional violence by men and women. The main findings are the distinct impacts produced by exposure to IPV on QoL according to gender. Only women had their QoL impaired when they were in situations of violence. Lower QoL scores were identified among older women who perpetrated violence against their partners, followed by those involved in bidirectional violence, and lastly among those who suffered aggression by their partners. But men did not have their QoL affected when perpetrating or suffering violence.

The older people in Florianópolis showed high QoL (men =46.8; women =45.8) when compared to population-based studies carried out in other countries¹⁶,²⁴ using the same measurement instrument (CASP-19). In Ireland, the QoL of older people was lower (43.8), as well as in England (42.5). The average QoL scores were similar between genders in the current study, differing from other population surveys¹⁶,²³. Said results may be related to the fact that the present study includes only older people with partners who had a high level of QoL when compared to non-participants. Older people married or with a partner had a higher QoL than single women. The partner’s presence is considered positive when there is social support and personal appreciation in the relationship²⁵.

It was found that women had their QoL impaired in situations of violence, which remained impaired even after the analysis adjusted by the factors of age group, income, cognitive deficit, depression, and dependence on ADL, while for men the QoL remained unchanged regardless of whether they suffered or committed such acts. This finding shows that the impact of violence on QoL differs significantly between older women and men. Gender differences may be the result of disparities in health and the different ways in which older women and men respond and deal with health problems and adverse experiences. In our culture, the male and female roles contribute to conflicts between couples and are often associated with violence in the relationship²⁶,²⁷. The binomial masculinity and violence has traditionally been understood as if the second term was part of the first. In this scenario of association between being male and being violent, gender relations are built and reproduced, thus legitimizing violence as a reference to distinguish men from women²⁸. This construction corroborates the differences in the impact of IPV on QoL between genders found in the present study.

For older women, both committing and suffering violence result in lower QoL scores. A greater negative impact on QoL is emphasized for women who perpetrate violence against their partners (-3.15; 95%CI: -4.84; -1.45) when compared to those who suffered the aggression (-1.62; 95%CI: -3.06; -0.17). It seems paradoxical that older women who perpetrate IPV have their QoL more compromised compared to those who suffer it since literature²⁶,²⁷ shows that serious consequences of violence are related to women being the victims. The suffering caused can work as a catalyst for installed or predisposed illness processes such as impaired physical health, dependence on activities of daily living, cognitive deficit, and a greater likelihood of mental health problems such as anxiety and depression, and social isolation in older age²⁸,²⁹ in addition to lower self-perceived health³⁰ which compromise the QoL of people involved in violence.

Note that although collecting data on perpetrators is challenging, understanding the role of factors such as socioeconomic status as the perpetrator’s financial dependence is an important barrier to seeking help, others such as mental health and abuse substance abuse as well as dependence and interdependence between perpetrators and victims require attention since understanding these characteristics is important to predict and prevent abuse of the older people²⁷,³¹.

In the present study, the role of women in IPV situations affects both QoL domains - control and autonomy, personal fulfillment and pleasure. Note that women are constantly held responsible for the care and success of relationships, being more likely to feel guilty for their partner’s violent behavior, which impacts their well-being. The expectation that women should be caregivers, sensitive, and conciliatory in the family environment not admitting that they become aggressive and violent ³² results in their QoL being even more compromised when they play the role of IPV perpetrators. Some authors³³,³⁴ describe the perpetration of IPV by women as a defense reaction against aggressions they previously suffered. Adopting violence to resolve marital conflicts evidences the lack of other resources to mediate problems, exposing the vulnerability of the relationship. In line with these results, an American study²⁷ found an association between low self-esteem and perpetration of this violence only for women.

It should be noted that IPV among older people is nothing new emerging in this age group. A population-based study carried out with adults in Florianópolis³² found that the prevalence of suffering physical violence was high for both men (17.5%) and women (16.1%), but with negative consequences in greater levels on women’s health, evidencing that violence in intimate relationships and their effects is maintained throughout life. An Australian longitudinal study stated that IPV had a lower QoL for women over the 16 years of study, perpetuating for three generations and explaining that the damage caused by violence to those who experience it is intergenerational³⁵.

It emphasizes the negative impact of IPV on the well-being of the older women interviewed, compromising positive and healthy aging and pointing to the need for strategies to prevent this form of violence at this stage of life. Women in situations of violence are more susceptible to remain socially isolated, away from friends and family, remaining restricted to the domestic environment, and living with the perpetrator of violence³¹. The consequences of IPV are hugely impacting the health and happiness of those involved, extending and affecting the well-being of families and even entire communities⁷.

Although the prevalence of IPV in older people shows gender symmetry³³, the effects produced by exposure to violence QoL are asymmetric since being in a situation of IPV did not have any impact on men’s QoL. However, this impact on women’s QoL is permeated by a gender inequity that must be fought since acts of aggression, abuse, and humiliation in the relationship are socially tolerable³⁰.

Based on the gender theory²⁹, men being violent against their partner corresponds to the correct way of behaving, a historically constructed and valued way to solve problems. The violence perpetrated by the partner remains naturalized and trivialized, being often not even considered as such³⁴. Moreover, violence is usually recognized by men only in situations of public life: in urban, impersonal, and/or anonymous relationships. Aggressive behaviors and acts of violence in the family and conjugal spheres are considered the role of the head of the family, therefore relationship violence remains invisible and neglected³⁴.

To reduce the IPV on men and women, it is necessary to deconstruct the hierarchical roles of gender and the reduction of structural factors supporting these inequalities, which are certain interventions of great value to cope with violence and the achievement of healthy aging, promoting QoL among the older people, especially women who are most affected³⁵,³⁶. Acknowledgment of situations of intimate relationship violence among older people in healthcare services is still an incipient practice in the Brazilian context³⁰. There are services to care for older people who are victims of abuse and mistreatment, but a larger part of this population may be suffering the consequences of IPV in their relationships.

Regarding the QoL, a specific definition transcending health-related issues and measuring the aspects of subjective perception of life was adopted for the older population. The analysis of a representative population-based sample of a capital of southern Brazil carried out by validated instruments and high reliability – CTS-1 and CASP-19 – achieved reliable results allowing to conclude that women had their QoL impaired by IPV and that being a perpetrator had a greater negative impact than being a victim, while for men this form of violence did not alter their levels of QoL.

The age profile of the sample is among the limitations found, which included older people aged 63 years and over. This group differs from the target population of the study which was planned to be applied to individuals aged 50 years and over. This limitation potentially implied a lower variability in the responses to the items, attenuating the magnitude of all the parameters estimated in the present study. Another limitation refers to the use strategy of CASP-19, which was face-to-face and not by self-response which may affect the responses to the items of the instrument by overestimating the QoL assessment resulting from the process. This type of administration was adopted in the study to allow even the older people with low levels of education or low visual acuity to answer the survey questionnaire.

Conclusion

The present study addressed the association of IPV on Quality of Life (QoL) of older women and men. An asymmetric impact was found between the gender and only women had their QoL impaired when experiencing situations of intimate partner violence (IPV), whether suffered, perpetrated, or bidirectional. Note that there are lower QoL scores for older women who have perpetrated IPV against their partners.

IPV has significant psychological and health impacts in older women, often exacerbated by the duration of violence. Prolonged abuse and essential social transitions such as children leaving home can lead to feelings of hopelessness, unfulfilled life expectations, and deep social isolation, reducing the quality of life of this population.

The importance of promoting policies to assist in the prevention of IPV of older people is emphasized, as this population faces barriers to be helped, in addition to being a topic that has not been investigated especially in terms of interventions. From the importance of QoL for the area of gerontology and the advance that CASP-19 represents – because it is aligned with new theoretical constructs on aging – it is expected that the present study will contribute to the notion of healthy and active aging and guide actions aimed at older populations in Brazil and worldwide.

The present study is relevant because it investigates a subject that has not been well investigated yet, addressing not only the violence suffered in the intimate relationship but also the perpetrated and the bidirectional ones. There is little research comparing the results of victimization and perpetration of IPV in men and women. Such conduct reduces the analysis bias since victims and aggressors are not previously defined, but such behaviors are analyzed in both sexes.

• No funding was received in relation to the present study.

REFERENCES

• ¹
  Machado DR, Kimura M, Duarte YADO, Lebrão ML. Violência contra idosos e qualidade de vida relacionada à saúde: estudo populacional no município de São Paulo, Brasil. Ciênc Saúde Colet. 2020;25(3):1119-28.
• ²
  Hackenberg EAM, Sallinen V, Koljonen V, Handolin L. Severe intimate partner violence affecting both young and elderly patients of both sexes. Eur J Trauma Emerg Surg. 2017;43(3):319-27.
• ³
  Castro VCD, Rissardo L K, Carreira L. Violence against the Brazilian elderlies: an analysis of hospitalizations. Rev Bras Enferm. 2018;71(Suppl 2):777-85.
• ⁴
  Hyde M, Wiggins RD, Higgs P, Blane DB. A measure of quality of life in early old age: the theory, development and properties of a needs satisfaction model (CASP-19). Aging Ment Health. 2003;7(3):186-94.
• ⁵
  Lima F, Hyde M, Chungkham H, Correia C, Campos AS, Campos M, et al. quality of life amongst older Brazilians: a cross-cultural validation of the CASP-19 into Brazilian-Portuguese. PLoS ONE. 2014;9(4):1-10.
• ⁶
  Cecchetto F, Oliveira QBM, Njaine K, Minayo MCS. Violências percebidas por homens adolescentes na interação afetivo-sexual em dez cidades brasileiras. Interface (Botucatu). 2016;20(59):853-64.
• ⁷
  Davies R, Lehman E, Perry A, McCall-Hosenfeld JS, Association of intimate partner violence and health-care provider-identified obesity. Women Health. 2016;56(5):561-75.
• ⁸
  Warmling D, Lindner SR, Coelho EBS. Prevalência de violência por parceiro íntimo em idosos e fatores associados: revisão sistemática. Ciênc Saúde Colet. 2017;22(9):3111-25.
• ⁹
  Ferdos J, Rahman M. Exposure to intimate partner violence and malnutrition among young adult Bangladeshi women: cross-sectional study of a nationally representative sample. Cad Saúde Pública. 2018;34(7):1-10.
• ¹⁰
  Coker AL, Follingstad DR, Garcia LS, Bush HM. Intimate partner violence and women’s cancer quality of life. Cancer Causes Control. 2017;28(1):23-39.
• ¹¹
  Universidade Federal de Santa Catarina. EpiFloripa: condições de saúde de adultos e idosos de Florianópolis. Florianópolis; 2009 e 2013/2014 [Internet]. São Paulo: UFSC; 2009 [acesso em 04 jun. 2020]. Disponível em: http://www.epifloripa.ufsc.br/
  » http://www.epifloripa.ufsc.br/
• ¹²
  Schneider IJC, Confortin SC, Bernardo CO, Carvalho CB, Antes DL, Pereira KG, et al. Estudo de coorte EpiFloripa Idoso: métodos, aspectos operacionais e estratégias de seguimento. Rev Saúde Pública. 2017;51:1-10.
• ¹³
  Programa das Nações Unidas para o Desenvolvimento. Relatório do Desenvolvimento Humano 2010: a Verdadeira Riqueza das Nações: vias para o desenvolvimento humano. Portugal: IPAD; 2010. Edição do 20º Aniversário.
• ¹⁴
  Instituto Brasileiro de Geografia e Estatística. Estimativas populacionais residentes, em 1º de julho de 2009, segundo os municípios [Internet]. Brasília, DF; 2009 [acesso em 04 jun. 2020]. Disponível em: http://www.ibge.gov.br/home/estatistica/populacao/estimativa2009/POP2009_DOU.pdf
  » http://www.ibge.gov.br/home/estatistica/populacao/estimativa2009/POP2009_DOU.pdf
• ¹⁵
  Kim GR, Netuveli G, Blane D, Peasey A, Malyutina S, Simonova G, et al. Psychometric properties and confirmatory factor analysis of the CASP-19, a measure of quality of life in early old age: the HAPIEE study. Aging Ment Health. 2015;19(7):595-609.
• ¹⁶
  Sexton E, King-Kallimanis BL, Conroy RM, Hickey A. Psychometric evaluation of the CASP-19 quality of life scale in an older Irish cohort. Qual Life Res. 2013;22(9):2549-59.
• ¹⁷
  Anacleto AJ, Njaine K, Longo GZ, Boing AF, Peres KG. Prevalência e fatores associados à violência entre parceiros íntimos: um estudo de base populacional em Lages, Santa Catarina, Brasil. Cad Saúde Pública. 2007;25(4):800-8.
• ¹⁸
  Reichenheim ME, Moraes CL, Szklo A, Hasselmann MH, Souza ERD, Lozana JDA, et al. The magnitude of intimate partner violence in Brazil: portraits from 15 capital cities and the Federal District. Cad. Saúde Pública. 2006;22(2):425-37.
• ¹⁹
  Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for clinician. J Psychiatr Res 1975;12(3):189-98.
• ²⁰
  Paradela EM, Lourenço RA, Veras RP. Validação da escala de depressão geriátrica em um ambulatório geral. Rev Saúde Pública. 2005;39(6):918-23.
• ²¹
  Rosa TE, Benício MH, Latorre MR, Ramos LR. Fatores determinantes da capacidade funcional entre idosos. Rev. Saúde Pública. 2003;37(1):40-8.
• ²²
  Netuveli G, Wiggins RD, Hildon Z, Montgomery SM, Blane D. Quality of life at older ages: evidence from the English longitudinal study of aging (wave 1). J Epidemiol Community Health. 2006;60(4):357-63.
• ²³
  Tampubolon G. Delineating the third age: joint models of older people’s quality of life and attrition in Britain 2002-2010. Aging Ment Health. 2015;19(7):576-83.
• ²⁴
  Bartlett BA, Iverson KM, Mitchell KS. Intimate partner violence and disordered eating among male and female veterans. Psychiatry Res. 2018;260:98-104.
• ²⁵
  Pengpid S, Peltzer K. Lifetime spousal violence victimization and perpetration, physical illness, and health risk behaviours among women in India. Int J Environ Res Public Health. 2018;15(12):1-10.
• ²⁶
  Barros CRS, Schraiber LB. Violência por parceiro íntimo no relato de mulheres e de homens usuários de unidades básicas. Rev Saúde Pública.2017;51:1-10.
• ²⁷
  Roberto KA, Hoyt E. Abuse of Older Women in the United States: A Review of Empirical Research, 2017-2019. Aggres Violent Behav. 2020:1-10.
• ²⁸
  Ho CS, Wong SY, Chiu MM, Ho R. Global prevalence of elder abuse: A metaanalysis and meta-regression. East Asian Arch Psychiatry. 2017;27(2):43-55.
• ²⁹
  Nunes DP, Brito TRPD, Duarte YADO, Lebrão ML. Cuidadores de idosos e tensão excessiva associada ao cuidado: evidências do Estudo SABE. Rev Bras Epidemiol. 2019;21(Supl.2):1-10.
• ³⁰
  Roberto KA, Hoyt E. Abuse of Older Women in the United States: A Review of Empirical Research, 2017-2019. Aggress Violent Behav, 2020:1-10.
• ³¹
  Pathak N, Dhairyawan R, Tariq S. The experience of intimate partner violence among older women: a narrative review. Maturitas 2019;121:63-75.
• ³²
  Lindner SR, Coelho EBS, Bolsoni CC, Rojas PF, Boing AF. Prevalência de violência física por parceiro íntimo em homens e mulheres de Florianópolis, Santa Catarina, Brasil: estudo de base populacional. Cad Saúde Pública. 2015;31(4):815-26.
• ³³
  Loxton D, Dolja-Gore X, Anderson AE, Townsend N. Intimate partner violence adversely impacts health over 16 years and across generations: a longitudinal cohort study. PloS ONE. 2017;12(6):1-10.
• ³⁴
  Yon Y, Mikton CR, Gassoumis ZD, Wilber KH. Elder abuse prevalence in community settings: a systematic review and meta-analysis. Lancet Glob Health. 2017;5(2):147-56.
• ³⁵
  Policastro C, Finn MA. Coercive control and physical violence in older adults: Analysis using data from the National Elder Mistreatment Study. J Interpers Violence. 2017;32(3):311-30.
• ³⁶
  Pathak N, Dhairyawan R, Tariq S. The experience of intimate partner violence among older women: A narrative review. Maturitas. 2019;121:1-10.

Edited by

Edited by: Maria Helena Rodrigues Galvão

Publication Dates

• Publication in this collection
  11 June 2021
• Date of issue
  2021

History

• Received
  28 Aug 2020
• Accepted
  10 Mar 2021