Psychosocial Links Between Internet Use and Mental Health in Later Life: A Systematic Review of Quantitative and Qualitative Evidence

Abstract

Research on the role of information and communication technology (ICT) use for active aging is limited. The aim of this systematic review is to investigate the link between Internet use and mental health among older adults. The review was conducted based on searches in 9 electronic databases (2002-2014). A meta-synthesis approach was applied, examining quantitative (18) and qualitative (14) studies. The findings from the synthesis of quantitative statistical data indicate an overall positive association between Internet use and mental health and its psychosocial covariates in later life. The psychosocial links between Internet use and mental health identified from the qualitative data were (a) enhanced interpersonal interaction at individual level, (b) increased access to resources within the community, and (c) empowered social inclusion at society level.

The results highlight the multi-level psychosocial links between Internet use and mental health, which may be applied in initiatives targeting healthy aging in various settings.

Keywords

older adults mental health Internet usage psychosocial factors systematic review

Introduction

In recent years, there has been increased interest in research looking at the use of ICTs and its influence on both the individual and society. Because Internet use is more common among younger people (Pew Research Center, 2013; Seybert, 2012), ICT-related research has predominantly targeted these age groups, while older adults have received less attention in this area.

In addition to being overlooked in Internet-related research, the aging population is a group often disregarded in mental health research (Forsman, Ventus, van der Feltz-Cornelis, & Wahlbeck, 2014). Mental health and well-being of older adults is a public health concern, as mental health problems and disorders are highly prevalent in later life (World Health Organization [WHO], 2013). The need to prioritize “adding life to years instead of years to life” (FUTURAGE, 2011) has been emphasized, highlighting the psychosocial (emotional and social; Bowling, 2007) aspects of health alongside the physical facets in an increasingly aging population (European Commission, 2013). There is a need to explore new forms and possibilities for supporting mental health in older people. The growing evidence on ICT use and its impact on mental health need to be further examined—also among the older age groups.

A previous review of quantitative studies examining older adults’ ICT use and mental health (van der Wardt, Bandelow, & Hogervorst, 2012) presented inconclusive results revealing both positive and negative associations to cognition and well-being. More research with a psychosocial approach is warranted, further examining the links between various aspects of mental health and Internet use.

The present review attempts to summarize the evidence base of correlational data on Internet use and related psychosocial factors and its links to mental health among older adults, utilizing available data from both quantitative and qualitative studies.

Method

This literature review applies a mixed-method approach as it includes quantitative and qualitative research, as well as studies that combine the two methods. The evidence from the studies included was synthesized and interpreted with equal weight (Noyes & Lewin, 2011a, 2011b). This data triangulation allows for a more in-depth analysis, utilizing the strengths of each methodology and increasing the overall validity of the review (Harden, 2010; Hesse-Biber, 2010).

Inclusion and Exclusion Criteria for Studies

Types of studies

This review includes studies applying quantitative statistical analyses, as well as those using qualitative data methodologies, when studying the links between Internet use and various mental health outcomes. Studies reporting only frequencies of older adults’ Internet use for various purposes were excluded from this review. Studies only examining associations between older adults’ Internet use and socio-demographic factors or associations between Internet use and non-psychosocial factors were excluded. To be included in the review, the participants were required to be independent or assisted Internet users. Internet access points were not limited to personal or public computers but included other connected devices such as smart phones and tablets. Finally, studies only reporting computer use, without further clarification or information on whether Internet use was measured, were excluded.

Types of participants

Studies looking at older adults, defined as adults above the age of 60, were considered in the review. This age cutoff is based on United Nations (UN) definition of old age (UN, Department of Economic and Social Affairs, Population Division, 2002). Studies were included even if the age range of the participants began under 60, if the mean age of the sample was 65 or older, or if data were extractable separately for study participants above the age of 60. No upper age limit was applied. Studies failing to provide information on participants’ age were excluded from the review. Participants with various living arrangements and varying physical and/or mental health status were included in the review—however, considering the requirements of independent or assisted Internet use, most participants were expected to be relatively healthy older adults.

Types of outcome measures

The multidimensional concept of mental health encompasses functional as well as psychological and social components (WHO, 2014). The psychosocial factors related to mental health are the outcomes of interest in this review. The outcomes considered were directly (e.g., quality of life, depressive symptoms) or indirectly (e.g., functional independence, experienced loneliness) associated to mental health, while, for example, perceived benefits in service provision or various biological or neurological measurements (e.g., cognition) were not considered.

Data Collection—Searches

Nine electronic databases were searched for eligible studies: AgeLine, ASSIA, Cinahl, Medline, Sociological Abstracts, PsycINFO, Social Care Online, SocIndexfull text, and Web of Science (see the appendix for specific details on search strategies for the different databases). Aiming for a wide interdisciplinary approach, search strategies with broad search terms were applied. Books, book chapters, dissertations, and conference proceedings were excluded from this review. Research articles must have been published in English or have an English translation to be included. The searches had a time frame starting from January 1, 2002, to April 30, 2014.

Procedure and Analysis

All references retrieved from the searches were screened for inclusion by the first and second author independently. In unclear cases, consensus was reached by discussion. From the full-text publications, available data on study participants and outcome measures were extracted and coded independently by the authors. The reference lists of the selected publications were subsequently screened for further identification of relevant studies.

From studies reporting eligible quantitative data, correlational data on statistical associations between Internet use and mental health or psychosocial factors were extracted and analyzed (Table 1). The results of final analyses, for example, final models in regression analysis, were used when available. That is, the effects of potential confounders, such as socio-demographic factors, were simultaneously considered for the studied outcomes.

Table 1.

Description of Quantitative Studies (n = 18) Included in the Review.

Study Country	Methodology Setting	Sample	Type of Internet usage/Internet-related measurement	Mental health and/or psychosocial indicator	Association (+/−/0)	Methodological quality (+/−)
Berner, Rennemark, Jogreus, and Berglund (2012) Sweden (Parallel: Lagergren, M., Fratiglioni, L., Rahm Hallberg, I., Berglund, J., Elmståhl, S., Hagberg, B., . . . Wimo, A. (2004))	Baseline cross-sectional sample of SNAC, a longitudinal study Nationally representative data Stepwise logistic regression	n = 1,402 817 women, 585 men 60-96 years, M age = 85 years Respondents suburban living Majority of participants did not have a higher education Both Internet users and non-users	General/unspecified	Personality FFM Extraversion Openness SOC SOC, shortened version Life satisfaction LSI-A	0	Val +
						Gen +
					00	Rel +Obj +
Carpenter and Buday (2007)The United States	Cross-sectionalHierarchical logistic regression	n = 324230 women, 94 menMajority of respondents CaucasianLiving in a suburban, naturally occurring retirement communityComputer users M age = 75 years, non-users 80.5 years36% actively using computers	General, communication36% computer users (n =115)81% communication	Depression Geriatric Depression Scale, short form	0	Val +Gen −Rel +Obj +
				Loneliness (single question)	0
				Overall social impairment OARS Social Resources Rating Scale Total Score	−
				Functional independence Items from the Duke Older Americans Resources and Services Survey	+
Choi and DiNitto (2013a)	Data from the first interview wave of the NHATS	n = 6,680		Depression Two items	−	Val −Gen +Rel +Obj +
				Anxiety Two items	−
The United States	Nationally representative sample of U.S. Medicare beneficiaries aged 65+Face-to-face individual interviewsLogistic regression analyses	Respondents residing in their own home or another’s home (those in, e.g., residential care settings excluded)56% female81% Caucasian51% of the sample had a college degree or a higher level of education	Four types of Internet use: Health-related tasks, shopping/banking, email (with or without other types of Internet use), and other tasks	Self-efficacy Four items	+
				Social capital
				Volunteering during last month	+
				Visited family/friend last month	+
				Attended religious service last month	+
				Attended club/class last month	+
				Went out for enjoyment last month	+
				ADL/IADL impairment	0
Choi and DiNitto (2013b)The United States	Cross-sectionalMultinomial logistic regression analysis	n = 76370% female42% Caucasian, 36% African American, 21% HispanicOlder adults receiving home-delivered mealsRespondents low-income, disabled, and homebound17% current Internet users	General	Depression Self-reported (confirmed by case worker)	+	Val −Gen −Rel +Obj +
				Anxiety Self-reported (confirmed by case worker)	0
				ADL impairments Number of impairments	0
				IADL impairments Number of impairments	0
				Severe mental illness (not included in logistic regression analysis)
Cotten, Ford, Ford, and Hale (2012)The United States	Cross-sectionalRegression analysis and propensity score methods	n = 7,839M age = 73.5 yearsRetired adults 50 years or older, nursing home residents excluded	General	Depression CES-D	−	Val +Gen −Rel −Obj +
Cresci, Yarandi, and Morrell (2010)The United States	Data gathered as part of the 2001 Detroit City-Wide Needs Assessment of Older Adults (1,310 telephone calls, 100 face-to-face interviews)Target population consisted of all non-institutionalized persons 60 years or older.Logistic regression analysis	n = 1,410A probability sample of residents was selected.M age = 71.6 years995 females, 415 males83% African American13% Caucasian65% of the sample had an education level of high school or lower27% (386) computer users73% (1,024) computer non-users	General, computer and Internet usage	Mental health SF-12 MCS	0	Val +Gen +Rel −Obj +
				Social activity
				Active leisure time activities (requiring individuals to leave their home)	0
				Membership in community organizations or neighborhood groups	+
				Doing volunteer work	+
Erickson and Johnson (2011)Canada	Cross-sectional, non-random sampleCorrelations, multiple regression analyses	n = 122Community-based adults81 females, 41 malesAge: 60 years or above52% reported having started or completing a university or college educationRecruited from senior centers, education programs, and community organizationsMay represent only high-functioning older adults	FrequencyExtent of use for different purposes: Communication, information access, entertainment57% of the sample used the Internet every day25% never used the Internet	Loneliness UCLA Loneliness scale	0	Val +Gen −Rel +Obj +
				Life satisfaction LSI	0
				Self-efficacy 18 items from Self-Efficacy Scale	+ For total Internet use+ communication+ information usage
				Social support Social Support Appraisal Scale	0
				Depression BDI	0
Heo, Kim, and Won (2011)The United States	Cross-sectionalPearson correlation coefficients, t tests	n = 103Participants at a local senior center69 females, 34 malesM age = 70.5 years76% of participants Caucasian, 15.5% African American59% of participants had a high school education96% of respondents had a high school degree or higher level of education	Importance of Internet use for the individual	Leisure satisfaction (LSS)6 subscales Psychological Social Relaxation Educational	++++	Val +Gen −Rel +Obj +
		52 individuals in the high affinity group (perceived high importance of using the Internet)
James, Boyle, Yu, and Bennett (2013)The United States	Data from the Rush memory and Aging Project (longitudinal cohort study)Spearman correlation	n = 66160-101 years76% female91% CaucasianParticipants recruited from retirement and subsidized housing facilities	Frequency of Internet/email use	Depressive symptoms CES-D	0	Val +Gen −Rel +Obj +
				Loneliness Modified version of the Jong-Gierveld Loneliness scale	−
				Social support Four items from the Multidimensional Scale of perceived Social Support (significant other subscale)	+
				Social network size Total number of children, other relatives, and close friends seen at least once per month	0
				IADL disability Items from the Duke older Americans Resources and Services Project	−
Koopman-Boyden and Reid (2009)New Zealand	Cross-sectional, nation-wide random samplePhone survey dataChi-square test, t test	n = 1,68065-84 yearsMajority of respondents European/NZ EuropeanNon-institutionalized respondents51% used a computer to access Internet and email	Internet/email	Well-being		Val +Gen +Rel −Obj +
				WVS question	+
				WHO-QOL	+
				Satisfaction/dissatisfaction with 10 specific domains of well-being
				Health	0
				Education	0
				Work	0
				Economic standard of living	0
				Entitlements and rights	0
				Participation in leisure and recreation activities	+
				Physical environment inside and outside the house	0
				Personal safety	0
				Contact with family/contact with other people	0
				Cultural identity	0
Lam and Lam (2009)Australia	National, cross-sectional studyRandomized multistage samplingLogistic regression analyses	n = 784Care-givers aged 60 and above60% of respondents living in major cities38% of respondents had an education level of post-secondary, university, or above15.8% reported having used the Internet in the past 12 months61% had used the Internet for personal and private use, e.g., email, chat	GeneralPurposes of use	Mental health SF-12v2 Health Survey	+	Val +Gen +Rel +Obj +
Lelkes (2013)Europe (multiple countries)	Cross-sectionalESS datan = 50,000Nationally representative samples from 26 European countriesLogistic regression analyses	n = 11,000+65 years of age or above78% of respondents never use Internet or email for personal use	Email/Internet for personal useCompares respondents who are not Internet users to those who are online sometimes and those who are online every day	Social isolation		Val +Gen +Rel −Obj +
				Lack of a close friend	−
				Meeting socially less than once a month	−
				Feeling lonely	−
Medeiros et al. (2012)Brazil	Epifloripa Idoso, a cross-sectional household-based study on health conditions of older peopleResponse rate 89.2%	n = 1,6561,058 women, 598 men60-102 years85% of the respondents Caucasian	Ability to exchange online messages via computer (email, social networking site, instant messages, or voice messages).	Moderate/severe functional dependence ADL	−	Val +Gen +Rel +Obj +
	Poisson regression	23% had an education level of 12 years or moreM age = 70.4 years
Näsi, Räsänen, and Sarpila (2012)Finland	Cross-sectionalRandom sampleNationally representative survey Cultural Capital, Consumption, and Social networks among older adults (postal survey)Response rate 68%Logistic regression	n = 542302 women, 236 men60-79 years66% of respondents had a higher education43% frequents users (several times a week or daily)	Internet and/or email usage frequency	Leisure activity 20 leisure activities listed, indicated which were undertaken during past year		Val +Gen +Rel +Obj +
				Median 9 activities
				More active 9 activities or more	+
Stark-Wroblewski, Edelbaum, and Ryan (2007)The United States	Cross-sectionalOne-way MANOVAt tests	n = 298M age = 78.3 yearsThe majority of respondents CaucasianOlder adults attending a congregate meal service program	Email use20.1% of the respondents use email	Limitations in daily activities	−	Val +Gen −Rel +Obj +
				Feeling sad/blue	0
				Do things independently	+
				Volunteering is important	0
				Participating in activities	0
				Flexible daily coping	0
Sum, Mathews, Pourghasem, and Hughes (2008)	Cross-sectionalOnline surveyResponse rate 72%	n = 222Online adults above the age of 55M age = 69.7 years138 females, 83 males, 1 gender unknown	Internet Breadth ScaleUsing the Internet for:CommunicationFinding new peopleCommerceEntertainmentInformation	General Social capital		Val +Gen −Rel −Obj +
				Onyx Bullen Scale8 dimensions:
				Participation in community	+ Finding new people
				Feeling of trust	+ Communication
(Parallels: Sum, Mathews, Hughes, and Campbell, (2008) and Sum, Mathews, Pourghasem, and Hughes (2009))Australia	Spearman rank correlations, ordinary least-squares regression analyses	50% of respondents had completed some higher educationInternet most often used for communication		Neighborhood connections
				Family connections	+ History of Internet use
				Tolerance of diversity	+ History of Internet use
				Value of life	−Frequency of Internet use+ Communication
				Pro-activity in social context	−Entertainment
				Work connections	−Entertainment
				Well-being Australian Wellbeing Index
				Personal well-being	−for meeting new people, for entertainment
				National well-being	−Frequency of Internet use
				Personality (moderator) The Big Five Personality Test	− hours spent on the Internet
				Social loneliness SELSA
				Social loneliness	+hours spent on the Internet−communication
				Family loneliness subscale	+hours spent on the Internet+for meeting new people
				Romantic loneliness subscale	−hours spent on the Internet
				Sense of community
				Positive sense of offline community	+ Internet use for information seeking
				General sense of online community	+ Frequency of use+ communication
				Sense of alienation from offline community	− History of Internet use
Werner, Carlson, Jordan-Marsh, and Clark (2011)The United States(Parallels: Jackson et al., (2009); Clark et al., (2011))	Cross-sectionalData collected as part of baseline assessment for the Well Elderly II studyLogistic regression analyses	n = 460303 women, 157 menM age = 74.9 yearsCommunity-dwelling37% of respondents Caucasian, 32% African American, 20% HispanicRecruitment sites (21) included senior centers, senior housing complexes, and retirement communities23% reported computer use for email	EmailGeneral computer use, computer use for email23% reported computer use for email26% general computer use	Coping COPE subscales		Val +Gen −Rel +Obj +
				Active coping	0
				Religious coping	0
				Behavioral disengagement	−
				Mental disengagement	0
				Social support ISEL	0
				Perceived control Adapted version of the Perceived Control Scale	0
				Size of social network LSNS	0
				Psychological well-being SF-36 Mental composite
				Mental health	0
				Social functioning	0
				Role emotional	0
				Vitality	0
				Depression CES-D	0
				Life Satisfaction LSI-Z	0
Zhang and Umemuro (2012)Japan	Cross-sectional survey51% response rateCommunity-dwelling older adultsMultiple regression	n = 8445 men, 39 women61-86 years, M age = 73 yearsRespondents living in metropolitan and suburban areas	ICT hedonic and eudaimonic well-being5 online activities:shopping,news readingchattingonline gamesemailPerceived Usefulness Scale, ATCQ, CAS Scales	Daily life well-being PEAQ		Val +Gen −Rel −Obj +
				Daily life hedonic well-being	+
				Daily life eudaimonic well-being	+

Note. Methodological quality assessment criteria: Validity, generalizability, reliability, objectivity. Direction of associations: + = significant positive association; − = significant negative association; 0 = non-significant association. Methodological quality: + = fulfilled criteria, − = methodological weaknesses in this area. SNAC = Swedish National Study on Aging and Care; FFM = Five Factor Model (Swedish version); SOC = Sense of Coherence (Swedish translation); LSI-A = Life Satisfaction Index-A (version frequently used for older adults); LSI = Life Satisfaction index; NHATS = National Health and Aging Trends Study; ADL = Activities of Daily Living; CES-D = Center for Epidemiologic Studies Depression scale (eight-item version); SF-12 = the 12-item Short-Form Physical Health and Mental health Survey (MCS included in meta-analysis); MCS = Mental Component Summary; BDI = Beck Depression Inventory; LSS = Leisure Satisfaction Scale; WVS = World Value Survey; WHO-QOL = World Health Organization Quality of life Indicator; ESS = European Social Survey; SELSA = Social and Emotional Loneliness Scale; COPE = Multidimensional coping inventory; ISEL = Interpersonal Support Evaluation List; LSNS = Lubben Social Network Scale; SF-36 = the RAND-36-Item Health Survey, Version 2; LSI-Z = Life Satisfaction Index-Z; ICT = information and communication technology; PEAQ = Personally Expressive Activities Questionnaire; IADL = Instrumental activities of daily living ;

ATCQ = Attitudes Toward Computers Questionnaire; CAS = Computer Anxiety Scale; OARS = Older Americans Resources and Services; UCLA = University of California Los Angeles.

For qualitative interview data, a meta-aggregation approach was applied for data extraction (Pearson, 2004), which meant that only participant quotes included in the reviewed reports, and no author interpretations, were identified and included. Qualitative content analysis was chosen by the authors because of its step-by-step methodologically controlled approach, in comparison with other qualitative approaches, hence allowing for replication (Krippendorff, 2004). By using Krippendorff’s (2004) approach, the initial phase of analyzing the texts involved condensation by identifying and selecting the textual units throughout the data material (i.e., the sentences or paragraphs containing central aspects related to each other through their content and context). The condensation of the gathered qualitative data was undertaken after carefully and repeatedly reading through the gathered material, focusing on the research questions throughout the process. The analysis was further performed by systematically coding the condensed data, initially into preliminary themes or codes based on the textual units. An interpretive framework was created when initially reviewing quotes resulting in initial groupings under central themes. The framework was continually refined in the process (Noyes & Lewin, 2011b). All quotations relating to particular categories were grouped together and examined in terms of broader and overarching themes on a higher abstraction level. The grouping of the textual units (e.g., quotations related to the same central content) was an inductive process whereby various potential categories were rearranged and recoded to be mutually exclusive during the process of sorting quotations. Constant comparisons ensured that similarities and differences between the emerging categories were explored. The researchers discussed the analysis—the coding, categorizing, and the interpretation of the findings—throughout the work process to gain a mutual agreement. Illustrative quotations from the data material, representing the analyzed categories, have been included in the article to depict links between Internet use and mental health as subjectively perceived by the respondents.

The two types of data used for the review contributed equally to the analysis. They supplemented each other well as the data derived from qualitative studies provided a more in-depth illustration of older adults’ subjective interpretation of their experiences, whereas the quantitative data added more power to the synthesis due to larger sample sizes and provided statistics on the studied associations.

The final result of this multi-step analysis process is a meta-synthesis, a comprehensive presentation of the identified links between different mental health aspects and Internet use (Harden, 2010).

Study Flow

The searches yielded 5,539 hits, including duplicates, from different databases (see Figure 1). By removing duplicates and screening the retrieved titles and abstracts according to the review criteria, the number of publications was reduced to 306 studies that were subsequently screened in full-text format. The studies were divided into two groups: association studies (presented here) and intervention studies (to be presented in a forthcoming publication). Following a thorough examination of the publications, the final number of included association studies was narrowed to 32 studies: 18 and 14 studies provided quantitative and qualitative data, respectively. The main reasons for excluding studies from the review were the mean age of participants being too low or the measured outcomes not being psychological or social.

Figure 1.

PRISMA 2009 flow diagram.

Study Characteristics

The quantitative studies originated mostly from Western countries, and the number of respondents in the studies varied between 84 and 11,000. Some of the quantitative studies focused on Internet use in general, while others concentrated on communication or another more specific feature of Internet use. In some studies, the sample consisted exclusively of older Internet users, while in other studies, users and non-users were compared.

The number of informants in the studies providing qualitative data varied between 7 and 1,381, and the distribution of countries resembled that of the quantitative studies. Participant quotes without relevant content and extracts not including full quotations in author interpretations were excluded, leaving a total of 232 quotes included in analysis. Seventeen quotes are cited in the “Results” section, representing the views and experiences most frequently presented in the reviewed reports.

Quality Assessment of Included Studies

Four criteria were utilized by the two researchers independently to assess the quality of the review studies. The quantitative studies were assessed with the following criteria: validity (adequate measurement and study design), generalizability (the applicability of the study results to other settings), reliability (the consistency of reporting, for example, whether the process of research is logical, traceable, and clearly documented), and objectivity (neutrality, the extent to which findings are confirmable through the analysis being grounded in the data). The relatively equivalent criteria for the qualitative studies were credibility, transferability, dependability, and confirmability (Hannes, 2011; Patton, 2002). Few studies exposed shortcomings in more than one area (e.g., convenience sampling or scarce reporting).

Results

Results From the Quantitative Data Synthesis

The studies providing statistical data can be viewed in Table 1. Of the 18 studies, only 1 study showed no significant associations between Internet use and the psychosocial factors of interest (Berner, Rennemark, Jogreus, & Berglund, 2012). The remaining studies found at least one significant association between Internet use and the psychosocial outcomes reported. It should be stressed that most studies included several different measures of mental health or related psychosocial factors, and 42 of the 101 included psychosocial outcomes reported in the studies were non-significant.

The wide range of psychosocial outcomes and measurements used in the studies should be noted. Outcomes found in multiple studies were, among others, depressive symptoms (seven studies), experienced loneliness (five studies), and activities of daily living (ADL)/ instrumental activities of daily living (IADL) impairment (four studies, see Table 1). The type of Internet use considered in the included studies varied. While some considered Internet use in general (Berner et al., 2012; Choi & DiNitto, 2013b; Cotten, Ford, Ford, & Hale, 2012; Cresci, Yarandi, & Morrell, 2010; Lam & Lam, 2009), other studies differentiated between various types of Internet use or focused on a particular aspect of Internet use.

The statistically significant findings revealed that general well-being was positively associated to Internet use in one study (Koopman-Boyden & Reid, 2009), while daily life well-being was positively associated to Internet use in Zhang and Umemuro’s (2012) study. One study with a general mental health outcome showed a positive association to Internet use (Lam & Lam, 2009). Among the three studies showing significant associations between Internet use and a depression outcome, two studies indicated a significantly lower level of depressive symptoms among Internet users (Choi & DiNitto, 2013a; Cotten et al., 2012), while one study implied reversed results (Choi & DiNitto, 2013b). One of the studies reported a significantly lower level of anxiety symptoms among Internet users compared with non-users (Choi & DiNitto, 2013a).

Internet users scored significantly higher on self-efficacy outcomes than non-users in two studies (Choi & DiNitto, 2013a; Erickson & Johnson, 2011), and one study which looked at different aspects of coping revealed a lower level of behavioral disengagement among Internet users (Werner, Carlson, Jordan-Marsh, & Clark, 2011). Regarding ADL/IADL impairments, Internet users scored significantly lower than non-users in two studies (James, Boyle, Yu, & Bennett, 2013; Medeiros et al., 2012). Another study found a significantly lower level of limitations of daily activities among Internet users and a positive association between Internet use and doing things independently (Stark-Wroblewski, Edelbaum, & Ryan, 2007). Functional independence was positively associated to Internet use in Carpenter and Buday’s (2007) study.

Regarding social network and relationship outcomes, one study showed a higher level of social capital among older Internet users, specific items measuring, for example, volunteer activities and visits with friends or family (Choi & DiNitto, 2013a). Likewise, positive associations to Internet use were found in other studies in regard to social support (James et al., 2013), while Internet users experienced significantly lower levels of social isolation (Lelkes, 2013), loneliness (James et al., 2013), and overall social impairment (Carpenter & Buday, 2007). Two different aspects of engaging in social activities were positively associated to Internet use in one study (Cresci et al., 2010). In one study, a significant positive association to being involved in a higher number of leisure activities was found for Internet users (Näsi, Räsänen, & Sarpila, 2009), while Heo, Kim, and Won (2011) found a positive association between Internet use and four types of leisure satisfaction. Satisfaction with leisure and recreation was also found to be greater among Internet users in the Koopman-Boyden and Reid (2009) study.

Due to several studies looking at Internet use in general and not specifying Internet use function, it is difficult to distinguish differences in associations according to the type of Internet use examined. One study did, however, look at different aspects of use for different outcomes (Sum, Mathews, Pourghasem, & Hughes, 2008). Different purposes of Internet use (finding new people, entertainment, commerce, communication, information) along with Internet use frequency and Internet use history were surveyed in this study. Using the Internet for communication purposes, and for finding new people, was associated with higher levels of social capital among respondents, while being online for entertainment purposes showed a reversed effect. Well-being was lower among respondents using the Internet for meeting new people and for entertainment purposes. Loneliness was lower among respondents using the Internet for communication, while respondents using the Internet for finding new people showed higher levels of loneliness than non-users. Finally, Internet use for information seeking and for communication was associated with a stronger sense of community.

Results From the Qualitative Data Synthesis

The evidence from the qualitative data material is presented below, conveying the older participants’ perceptions of psychosocial links between Internet use and mental health. Details on the included studies can be viewed in Table 2. The emerging results are grouped under three themes: (a) new ways of communicating and connecting, (b) an increase in resources according to needs and preferences, and (c) mastering new technologies as a means of increased social inclusion. Illustrative quotes are presented within the identified themes, followed by a reference to the original publication.

Table 2.

Description of Qualitative Studies (n = 14) Included in the Review.

StudyCountry	Data collection and analysis	SampleSetting	Internet usage/Internet-related measurement	Emerged themes	Methodological quality (+/−)
Aguilar, Boerema, and Harris (2010)Australia	Two online focus groupsInterpretative phenomenological analysis	n = 93 women, 6 men65-82 yearsLiving in both rural and urban areas in various Australian statesUsing a computer for between 2 and 20 years	General/unspecifiedExperiences of computer use	Five central themes emerged—meanings the participants attribute to computer use Feeling in control An important part of daily life Keeping the brain active Relating to others Personal gains from usage	Cred +Transf +Depend +Conf +
Beckenhauer and Armstrong (2009)The United States	Single, face-to-face, semi-structured interviewsConstant comparative methodParticipants selected from Phase I of a mixed-method study exploring normative aging, communication, and quality of life in older adults	n = 1413 women, 1 man61-85 yearsmajority of participants community-livingWell-educated sample (selected from a university cultural/educational organization, and a teacher retirement center)Owning a computer or web televisionParticipants used the Internet an average of 7.35 hr per week	General (but focus on communication)	Central themes: Communication changes Changes due to normative aging processes Changes due to advancements in communication Communication mediums Convenience Appropriateness Relationship Personal choice Computer ownership differences Logistics Effect on other mediums Services available Social network The role of communication in quality of life	Cred +Transf +Depend +Conf +
				Connecting vs. isolation Loss of communication Quantity and quality of communication Dissatisfaction regarding communication Time Lack of connection Physical limitations Improvements to dissatisfaction with communication Advances in technology as a concern
Cresci and Novak (2012)The United States	Focus group discussionsKey concept and emergent-theme technique	n = 1816 women, 2 men56-95 years old, M age = 72 years50% had not completed high schoolBoth Internet users and non-users among respondents	General	Physical healthMental, social, and spiritual healthInternet abilities and resources	Cred +Transf −Depend −Conf +
Gatto and Tak (2008)The United States	Paper survey questionnaire with open-ended questionsConvenience sampleResponse rate 52%	n = 5837 women, 21 men59-85 yearsMajority of respondents CaucasianMean 15.5 years of schooling	General/types of activitiesShort answers and comments reported here	Barriers and benefits of Internet use Benefits: connectedness satisfaction utility positive learning experience	Cred +Transf −Depend +Conf −
		Members of SeniorNet, a computer learning centerInternet users for more than a month55% used the Internet daily		Barriers: frustration limitations mistrust time issues
Khvorostianov, Elias & Nimrod (2012)Israel	Semi-structured in-depth interviewsConvenience sample	n = 3215 women, 17 menM age = 76 yearsmost participants (29) had an academic degreeall participants living in poor neighborhoodsJewish immigrants from the former Soviet Union in IsraelActive Internet users	General	Internet uses: managing health nurturing professional interests maintaining and extending social networks appreciating the past enjoying leisure	Cred +Transf −Depend +Conf +
Larsson et al. (2013)Sweden	InterviewsSnowballing recruitmentConstant comparative method	n = 1066-82 years6 women, 4 menBoth Internet users and non-users included	Performance of various Internet-based activities	To be an established performer Continuously developing the performance of meaningful IBAs Encouraged to perform IBAs by others and access to meaningful activities Participate in society in accordance with own preferences	Cred +Transf −Depend +Conf +
				To enter the role of a performer Preparing for the performance of IBAs considered as meaningful Encouraged by others and meaningful IBAs to potentially perform IBA Experience restricted participation in societyTo be a nonperformer Contacting others to perform the requisite IBAs An unidentified meaning associated with performing IBAs Feel pressure to increase participation by use of IBAs
Malta (2007)Australia	Online surveySemi-structured online interviews	n = 7Self-selected sample6 women, 1 man6 participants previously divorced, 1 widowedM age = 71 yearsHighly educatedHighly familiar with the Internet (using the Internet for between 7 and 20+ years)Mean daily Internet use 5.4 hr	Using the Internet to meet potential romantic partners, through online discussion groups or online dating or having consequentially met someone through online discussion forum	Online romantic relationshipsOnset and durationInstigation and development of relationshipCyber-flirting, cyber-sex and cyber-cheating	Cred +Transf −Depend +Conf −
Pfeil, Zaphiris, and Wilson (2009)The United Kingdom	Semi-structured interviewsTwo first groups interviewed face-to-face, third group interviewed via email or via private messaging	n = 3121 women, 10 menM age = 69.8 years10 non-internet users10 Internet users who use emailBoth groups recruited from senior centers11 Internet users that use online communitiesThese participants recruited from online communities	Different for the two Internet using groups:Support via email/support in online communities	Self-disclosure Trust Courage Degree of self-disclosureDeep support Emotional support Misunderstandings and misbehavior Advice SimilarityLight support Casual talk Keep in contact Humor and encouragementCommunity building Group feeling Prejudice about online communities Access to a high number of peopleInformation/facts Distribution of information Validation of information Information based on experience	Cred +Trans +Depend +Conf +
H. Russell (2011)Australia	Individual semi-structured interviewsHermeneutic phenomenological methodologyInterviews at least twice a year over a period of 7 yearsInitial interviews took place during the period of learning. Participants were further interviewed when they had ceased lessons.	n = 1611 women, 5 men63-86 yearsICT learners, participants undertaking lessons with voluntary peer tutor (on average once a week for 12 months)	General	Ontological and existential themes of being, becoming, and belongingBeing Personal achievement Self-confidenceBecoming Adaptive capacity Autonomy Using tools interactivelyBelonging Quality of relationships Continued participation Interest in life generally	Cred +Trans +Depend +Conf +
C. Russell, Campbell, and Hughes (2008)Australia	Self-selected sampleRecruited via online search engines and online interest groups for older people, electronic survey30 of 154 Internet users participated in second stage online qualitative data collection	n = 3055 to 89 yearsLarge majority Australian bornParticipants had been Internet users for a mean of 3.3. years	General	Relationships with informal networks of family and friends Bonding social capitalFormal ties to voluntary associations Bridging social capitalNature, frequency, perceived quality of locally based informal and formal contacts, and activities 11-item version of the Duke Social Support Index	Cred +Trans +Depend +Conf +
Sayago and Blat (2010)Spain	3-year ethnographical study of email useIn-situ observations and conversations with participants 2-3 times per week while participants attended course workshops and meetings	n = 388No age range or mean age reported, all respondents quoted were, however, above 60 years of ageParticipants had left school at age 16 the latestFairly representative of relatively deprived older people in a developed country	Email use	Nature of use Social circles Frequency of use and contact Geographical distance and its relationship with other means of contact Emailing and other daily activities Socialization, emotion, and accomplishmentInteraction barriers Inappropriate and excessive functionality Managing attachments and emails: making email their own Do not hurry Using the keyboard and the mouse: Alternative devices	Cred +Trans +Depend +Conf +
Sourbati (2009)The United Kingdom	Observational methods and interviews	18 tenants, (6 care professionals, this info not included in meta-analysis)No age range or mean age reported, allrespondents quoted were, however, above 60 years of age	General	Getting to know about the Internet: Encounters, experiences, and expectationsAssessing welfare service information online	Cred +Transf +Depend +Conf −
		2 sheltered accommodation complexes; extra care housing complex introducing networked computers (18 months earlier), and another facility not making available communal Internet accessInternet experience varied from none to having been online at least once
Whitbourne, Ellenberg, and Akimoto (2013) The United States	Online sample of 10,308 respondentsSelf-selected sample, respondents recruited via a blog	n = 1,38160 years and aboveAll respondents playing an online gameThe majority of respondents had less than a bachelor’s degree (68%)	Respondents playing a free online game	Reasons for video game playing Challenge Stress relief Try to beat friends and teammates Enjoy graphics and effectsBenefits of video game playing Feeling sharper Memory improvement Ability to see patterns have improved Can perform other timed tasks more quickly Feeling more confident	Cred −Transf −Depend +Conf +
Xie (2007)The United States	Semi-structured open-ended interviewsGrounded theorySnowballing technique	n = 209 women, 11 men67-92 years, M age = 76.8Almost all participants (19) highly educatedMembers of a senior-oriented computer club, SNUG	Online interaction through, e.g., email use, instant messaging, and online forums.	Online interactions with family, friends, or other members of the SNUG, or other SeniorNet membersLack of online interactions and relationshipsExchange of information in the offline settingCompanionship in the offline setting	Cred +Transf −Depend +Conf +

Note. Methodological quality: + = fulfilled criteria; − = methodological weaknesses in this area. Methodological quality assessment criteria: Credibility, transferability, dependability, confirmability. IBA = Internet-based activity; ICT = information and communication technology; SNUG = SeniorNet Users Group.

New ways of communicating and connecting

The participants of the qualitative studies described how electronic communication (emailing being the primary form of communication reported) helped maintain and increase contacts with family, friends, and others in their existent social network. The improved and more frequent communication appeared to have a beneficial effect on the quality of the relationships, which according to the participants affected their subjective experience of mental health:

I think that my quality of life has been better because of being able to be close to my children, and to my family because that is a powerful pull. . . . I have always had a warm feeling for family and this [computer and e-mail] has allowed me to continue that. (Beckenhauer & Armstrong, 2009, p. 839)

My health and happiness have been enhanced because of my ability to contact the people who are important in my life. (Aguilar, Boerema, & Harris, 2010, p. 31)

Furthermore, for established older Internet users, losing this newfound communication option was regarded as a threat to mental health as it held the potential to lead to loneliness:

I think that my life is enriched by the people I know, the people I communicate with, and I can keep up with their lives. And it would be very, very lonely I think if really all I had to depend on was the telephone or a letter because having instant communication makes a lot of difference. (Beckenhauer & Armstrong, 2009, p. 836)

For participants experiencing limitations in their everyday lives related to the aging process, online communication could be invaluable with regard to maintaining social contacts and social relationships:

My wife had a stroke last year and now cannot be left alone and this has restricted my outside activities . . . Since my wife became so dependent it has resulted in me being able to keep in touch with many friends that I would otherwise find it difficult to keep in contact. (C. Russell, Campbell, & Hughes, 2008, p. 80)

Moreover, many quotes illustrated how the Internet can not only enhance already established relationships but can also increase the individuals’ social network through new relationships formed online. The importance of these new contacts emerged in several of the studies, highlighting the added value they brought to the social networks and everyday life of the older individual:

I believe some friends I have made online know more about me—who I am and what I believe and feel—than my bridge buddies ever will. (Pfeil, Zaphiris, & Wilson, 2009, p. 163)

As illustrated in the following quote, online groups can provide a valuable source of support, especially if the individual might be lacking this type of support in his or her existent offline network:

The Bosom Buddies [health related online community] were sympathetic and supportive because they had all had breast cancer, and knew what I would be going through—physically, mentally, and socially. (Pfeil et al., 2009, p. 166)

According to the theoretical framework advocated by Lahtinen, Lehtinen, Riikonen, and Ahonen (1999), social interaction is a prerequisite for experienced mental health and well-being. Based on the qualitative data analysis, it is suggested that ICT technology plays an important role in the interaction and communication between the older individual and members of his or her social network. Furthermore, it seems that Internet use not only enables new opportunities for communication but also opportunities for connecting with new people. The facilitating role of the Internet for social interaction among older adults further underlines the link to mental health.

An increase in resources according to needs and preferences

Based on the synthesized qualitative data, the Internet provides and expands accessibility to resources other than those related to communication (e.g., public services, common interest groups). Internet use may support and enable participation in various activities and contexts chosen according to the older adults’ preferences, thereby having an enabling and empowering effect:

I love the scope of the world news and the ability to follow up on stories I see on TV. In other words, I feel sharper and therefore a person more in control of my life. (Aguilar et al., 2010, p. 29)

Overall, the participants of the included qualitative studies described how Internet activities contributed to a more active life on their own terms:

I always loved history. Now I read about everything I ever wanted to read: Ancient Egypt, China . . . Since I’ve got a computer, I almost never watch TV. The computer took all my spare time, but I don’t think I’m wasting it. On the contrary, I benefit from it. (Khvorostianov, Elias, & Nimrod, 2012, 593)

Being able to keep up a hobby or find a new interest can be an important feature of the Internet in a group where many individuals may find the increased spare time that retirement brings unfamiliar:

Genealogy is my main hobby, and the lists I belong to have opened up a whole new group of friends, and this has led to my husband and I, taking our caravan and joining other like-minded people at Cornish festivals, which I found out about on the Internet. (C. Russell et al., 2008, p. 81)

Through providing recreational opportunities and information resources in new ways, the Internet can promote mental health through factors such as an increase in leisure time activity and enhanced self-efficacy.

Another aspect, particularly expressed as relevant among widowers, is the Internet as a forum for meeting new partners. The participants pointed out the possibilities that e-based resources provide to those who would like to meet a new partner.

I had virtually no luck offline. First, I have no idea how to meet single men [because] I’d been married for a long time. . . . When I was in __ for six weeks, I had about 5 dates! (Malta, 2007, 94)

In addition, online sites and applications provide earlier unseen possibilities to reminiscence and revisit places earlier visited or lived in, an activity that has previously been revealed to have a positive impact on older individuals’ mental health (Bohlmeijer, Smit, & Cuijpers, 2003). Using the Internet for this purpose was especially mentioned as being important to older migrants and war veterans in the included studies:

I surf the websites of Volgograd and Leningrad. Because these are the places I miss the most. I look at the views of the city. I try to find those streets I know, images of houses that I am familiar with. This gives me a feeling of presence and kind of a sad happiness. (Khvorostianov et al., 2012, p.591)

I often use the Internet to get on to the Australian War Museum. Through that I have found a lot of my friends or relatives or people that I trained with and I can look up the Nominal Roll . . . it gives me satisfaction. Not because they got shot down but at least I know what happened to them and be able to use the computer and know where they are. (H. Russell, 2011, p. 119)

Mental health is strongly influenced by contextual factors (Bronfenbrenner, 1979; Lahtinen et al., 1999), such as accessibility to various resources in the community that are specific to the needs of individuals. Our review adds to this theoretical framework by integrating the virtual context and highlighting the resources accessible in the virtual community and the linked benefits for mental health among older people in particular. By identifying links—such as those illustrated above—between accessibility of Internet-based resources and mental health, the understanding of the psychosocial facilitators or barriers of active and healthy aging in a postmodern context is deepened.

Mastering new technologies as a means of increased social inclusion

The previous two themes emerging from the qualitative data material illustrate the potential ways in which Internet functions may support older individuals’ mental health. This third theme demonstrates the unique contextual circumstances surrounding the adoption of ICT technology in old age, due to, for example, limited technology experience or stereotypical attitudes (Cutler, 2005).

Older adults generally adopt new innovations at a slower pace than younger generations (Carey & Elton, 2010). In many cases, postmodern technology has not been previously used by older adults. Subsequently, adequate support regarding skill training as well as motivational support and encouragement is required.

The participants are often aware of their situation as non-users and the stereotypical perception of them as a group that has difficulties in adopting new technologies. This in itself has the potential of creating great pleasure when mastering this particular skill. Some participants expressed that entering the virtual world enhanced their mental health through promoting feelings of empowerment and improving self-esteem:

. . . Apart from being in touch with our nearest and dearest, and being more active, and all the things you can see here, we are teaching ourselves that we can still do things that people believe we can’t do because of our age. We are showing them and ourselves that we can. . . . (Sayago & Blat, 2010, p. 113)

Joysticks? I think that my grandchildren use them to play videogames, but for sending e-mails, they use the mouse. Imagine if they saw me using a joystick to send e-mails . . . I don’t want them to think their grandmother is frail. . . . (Sayago & Blat, 2010, p. 115)

The positive effects of experiencing oneself as a competent ICT user may be a reason for many participants wanting to be able to utilize standard technology devices, while others expressed a wish for simplified tools and applications.

However, realizing that acquired skills need continuous updating—as well as updating the actual technical devices themselves—seemed to be stressful and frustrating. This was combined with a fear of not being able to keep up with developments and the implications this could have for the individuals’ future independence:

I think the time will come fairly soon when the way people manage in the community will be dependent upon this technology and I want to remain as independent as possible. (Aguilar et al., 2010, p. 30)

I enjoy e-mailing and, in general, using computers, independently. When I started to e-mail, I was so reliant on other people that I got the impression I was useless. I didn’t want to bother people, you know. . . . (Sayago & Blat, 2010, p. 115)

Thus, Internet use can involve various concerns, regarding security issues or apprehension over what digital exclusion could imply for the older adult’s independence. These issues highlight problematic aspects that might, at least momentarily, be associated with perceived negative psychosocial factors. Nevertheless, as shown in previous research (Bryant, Corbett, & Kutner, 2001), the individual’s mental health and experienced well-being can benefit from challenging situations—if they feel they have the sufficient resources and abilities to meet them. This is an issue that needs further attention in terms of how to best support older adults in their Internet use and related needs.

To be socially included and up-to-date in today’s world means keeping up with the rapid advances in technology. In addition to the empowering effect that mastering the postmodern technology provides, another emerging psychosocial feature that seemed to contribute to mental health was the experience of social inclusion generated when crossing the digital divide. This experience of inclusion, both on a larger societal scale and on a smaller scale within the older adults’ own network was often emphasized.

. . . Yesterday, we had dinner with my children. They started talking about e-mailing, because we were talking about their work. They were very surprised when I broke in and started to use these words we use in class, like attachments and e-mail address. I could join in the discussion. I still remember when years ago I got so irritated that I asked them to change the topic of the conversation when we all met together in family events. Yesterday, I took part in the conversation, and that was very important for me. (Sayago & Blat, 2010, p.109)

These quotes illustrate the importance of being included in the ICT society, enabled by acquired ICT skills and knowledge.

As demonstrated above, adopting new technologies and crossing the digital divide are associated with different aspects of older adults’ mental health. If the individual receives adequate support, mastering technology can contribute to feelings of empowerment and capability. Furthermore, feelings of enhanced social inclusion in society, as well as among online peers, are associated to positive aspects of mental health among many older Internet users.

Discussion

The findings of this review show that there is an evidence base for links between Internet use and mental health and its psychosocial covariates in later life (as defined in this study). Based on the gathered findings from the 18 quantitative cross-sectional studies included in the review, it was concluded that all studies but 1 (Berner et al., 2012) found statistically significant correlations for at least one of the various mental health outcomes. These results suggest a relationship between Internet use and experienced well-being of the older individual.

However, it is important to recognize that a large number of measured outcomes were non-significant (42 of 101), and therefore, the results should be interpreted with caution. The inconclusive results are in line with previous review findings evidencing both positive and negative associations between older adults’ ICT use and well-being (van der Wardt, Bandelow, & Hogervorst, 2012). Regarding the few statistically significant findings pointing to an association between Internet use and lower levels of well-being (Sum et al., 2009) found in this review, they may be explained by the fact that these authors had distinguished between different aspects of Internet use. Follow-up studies could be useful for depicting associations between Internet use and mental health at different stages of the Internet use career of older individuals not identified in the cross-sectional data sets included here. The current review does, however, shed some further light on the topic by incorporating qualitative data in the analyzed data material using a systematic approach, which has not been done in previous reviews focusing on this theme.

Mental health should be viewed from a holistic viewpoint, including functional, mental, and social aspects. Consequently, mental health should be supported and protected—taking into account the broad range of determinants in different contexts and on various levels (WHO, 1986, 2005). Theoretically, the review results support the holistic perspective and highlight the importance of considering the psychosocial factors relevant to mental health and well-being both in research and in practice. Furthermore, the results confirm the relevance of the multi-level model of mental health by presenting three overarching psychosocial features available through Internet use in later life: (a) enhanced interpersonal interaction at individual level, (b) increased access to resources within the community, and (c) empowered social inclusion at society level. These findings support the use of an adopted version of Bronfenbrenner’s (1979) ecological model for the theoretical illustration of the dynamic interaction between the individual and his or her social context and the implications for mental health in later life (Forsman, 2012; Greenfield, 2011). According to this model, preferences, abilities, and attitudes at the individual level form an important basis for the individual’s well-being, at the same time as the social relationships at the interpersonal level, social contacts and resources at community level, and social participation at societal level are central psychosocial covariates of mental health in later life.

Furthermore, the review findings are in line with well-known theoretical concepts such as the functional model of mental health (Lahtinen et al., 1999) and the activity theory (Burbank, 1986; Havighurst & Albrecht, 1953; Lemon, Bengtson, & Peterson, 1972). The functional model defines mental health and well-being as a balanced interaction between the individual and their environment and is influenced by numerous factors on both micro and macro level: psychological and biological predispositions of the individual, social interactions in different social contexts, societal structures and resources, and cultural values in the community. The functional model of mental health shares several theoretical concepts with the activity theory (Burbank, 1986; Havighurst & Albrecht, 1953; Lemon et al., 1972), such as the view of the social context (both micro and macro level) and social interaction and participation as key components of mental well-being. This theory describes the psychosocial aging process and emphasizes the importance of maintained social activity and participation: to enhance mental well-being, the aging individual must substitute new meaningful social roles for those that are lost because of age (Havighurst & Albrecht, 1953; Lemon et al., 1972). This review illustrates the opportunities for active aging that the ICT technology can provide in today’s postmodern society.

Methodological Viewpoints

A potential source of bias of reviews in general is the search strategies used for the identification of relevant studies. Even though general and broad search terms were applied in this review, to capture a wide range of studies conducted, some specialized niches of research might be underrepresented in the search results. In addition, the exclusion of non-English language publications should be considered a limiting factor. It should be noted that many of the included studies were limited in terms of sample size and study power. Due to this, some of the studies included in the quantitative data material were deemed not generalizable and/or not reliable in the quality assessments. Also the risk of non-significant associations for specific outcomes not being included in study findings should be considered. Because a large number of non-significant associations were reported in the studies providing quantitative data, this perhaps increases the risk of a file drawer effect.

Furthermore, although many different countries were represented among the included studies, one limitation of the review was that the majority of studies originated from the Western world. In addition, although various sub-groups of the aging population (such as widowers, informal caregivers, individuals suffering from chronic conditions and with substantial care needs, ethnic minorities, and individuals with varying socio-economic status) were represented in the review, it is possible that other relevant sub-groups were not represented in the study material.

Another limitation, related to the synthesizing task of the review, was that a meta-analysis of the quantitative data material could not be performed due to the limited number of studies and various measurements used for the outcomes considered. Nonetheless, reviews covering both quantitative and qualitative evidence are rare and valuable for gaining an overall picture of a research area, despite the impossibility of ascertaining causality based on cross-sectional data. Furthermore, the inclusion of qualitative data analyzed in a systematic manner using a content analysis approach that enables replication should balance the limitations of the cross-sectional study design. This is because many quotes presented illustrate the participants’ perceptions of how Internet use was associated with various aspects of experienced mental well-being. It should, however, be noted that the quotes that the qualitative data synthesis is based on are restricted to the selection in the original articles.

Next Steps in Research and Practice

The review results highlight psychosocial links between Internet use and mental health that can be utilized in initiatives aiming to improve the everyday life of older adults. The Internet has revolutionized the possibilities of conducting widespread interventions with a psychosocial approach, focusing on supporting individuals’ mental health through initiatives with a social or psychological content (Forsman, Nordmyr, & Wahlbeck, 2011). There is, thus, a need for intervention-based research, with a focus on evaluating the psychosocial effects on older intervention participants, adding to earlier findings regarding the cost-effectiveness of e-based interventions (Knapp, McDaid, & Parsonage, 2011). For example, there is moderate evidence from several studies that computer/Internet training reduces levels of loneliness and enhances quality of life among older community-dwelling adults (e.g., Blazun, Saranto, & Rissanen, 2012; Shapira, Barak, & Gal, 2007). The effectiveness of controlled intervention studies, employing Internet use as a tool to promote mental health, is currently being evaluated in a meta-analysis by the authors of this review. These forthcoming research findings will shed further light on the causal mechanisms between mental health among older adults and ICT use identified in this review, and their functioning in an intervention context. Apart from evaluated interventions, more studies with a focus on following up older adults’ Internet use careers in a longer perspective and the eventual fluctuations in associations to well-being would be valuable. Finally, the usefulness of considering the identified mechanisms in the design and implementation of ICT education for different sub-groups within the older population could be further explored.

Footnotes

Appendix

Review Search Strategies, Searches Applied April 30, 2014.

Database	Search terms
Ageline
#1	AB (“older adult” OR “older people” OR “senior citizen” OR “frail elder” OR aging OR “in old age” OR “nursing home”) OR TI (elder* OR “older adult” OR “older people” OR “late* life” OR “late adulthood” OR “senior citizen” OR “frail elder” OR “very old” OR ag*ing OR “in old age”)
#2	AB (Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “web-based” OR “Internet-based” OR “Internet intervention” OR ICT OR “communications media” OR “information technology”) OR TI (computer* OR Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “web-based” OR “Internet-based” OR “Internet intervention” OR ICT OR “communications media” OR “information technology”)
#3	S1 AND S2
Social Care Online
	(freetext=“older people” and freetext=“internet”) and publicationdate>=20020000 and publicationdate<=20120000
Web of Science
#1	TI=(computer* OR internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “email” OR chat OR “e-resource” OR “e-intervention” OR “internet-mediated intervention” OR online OR “webbased” OR “internetbased” OR “internet intervention*” OR ICT)
#2	TS= (internet or “information technology” or “internet usage”)Databases=SCI-EXPANDED, SSCI, A&HCI Timespan=2002-01-01 - 2012-08-31
#3	TI= (elder* OR geriatr* OR “older adult” OR “older people” OR “late life” OR “late adulthood” OR senil* OR “senior citizen” OR “frail elder” OR “very old” OR aging OR residential)
#4	TS= (aging or older adults OR gerontology OR geriatrics)
#5	#2 OR #1
#6	#4 AND #3
#7	#6 AND #5
SocIndex fulltext
#1	TI (computer* OR Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “web-based” OR “Internet-based” OR “Internet intervention” OR ICT OR “communications media” OR “social media”) OR SU (Internet)
#2	TI (elder* OR geriatr* OR “older adult” OR “older people” OR “late* life” OR “late adulthood” OR senil* OR “senior citizen” OR “frail elder” OR “very old” OR aging OR residential OR “in old age”) OR SU (aging OR “older adults” OR “in old age”) OR AB (elder OR geriatr* OR “older adult” OR “older people” OR “late* life” OR “late adulthood” OR senil* OR “senior citizen” OR “frail elder” OR “very old” OR aging OR residential OR “old age”)
#3	(TI (elder* OR geriatr* OR “older adult” OR “older people” OR “late* life” OR “late adulthood” OR senil* OR “senior citizen” OR “frail elder” OR “very old” OR aging OR residential OR “in old age”) OR SU (aging OR “older adults” OR “in old age”) OR AB (elder OR geriatr* OR “older adult” OR “older people” OR “late* life” OR “late adulthood” OR senil* OR “senior citizen” OR “frail elder” OR “very old” OR aging OR residential OR “old age”)) AND (S1 and S2)
Medline
#1	exp internet/ or exp computer communication networks/ or exp social networking/
#2	(internet or computer or “computer technology” or “social networking” or web-based or online or “internet intervention” or blogging or “social media” or facebook or skype).ab.
#3	1 and 2
#4	exp aged/
#5	(elder* or geriatr* or “older adult” or “older people” or “late life” or “late adulthood” or “senior citizen” or “very old” or aging or residential).ab.
#6	4 and 5
#7	3 and 6
ASSIA
	ab((elder* OR geriatric* OR “older adult” OR “older people” OR “late adulthood” OR “senior citizen” OR “frail elder” OR “very old” OR aging)) AND ab((internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “social network” OR “email” OR chat OR “e-resource” OR “e-intervention” OR “e-service” OR “internet-mediated intervention” OR online OR “web-based” OR “internet-based” OR “internet intervention*” OR ICT OR “social media”))
Sociological Abstracts
	ab((elder* OR geriatric* OR “older adult” OR “older people” OR “late adulthood” OR “senior citizen” OR “frail elder” OR “very old” OR aging)) AND ab((internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “social network” OR “email” OR chat OR “e-resource” OR “e-intervention” OR “e-service” OR “internet-mediated intervention” OR online OR “web-based” OR “internet-based” OR “internet intervention*” OR ICT OR “social media”))
PsycINFO
#1	AB (“older adult” OR “older people” OR “senior citizen” OR “frail elder” OR aging OR “in old age” OR “nursing home”) OR TI (elder* OR “older adult” OR “older people” OR “late* life” OR “late adulthood” OR “senior citizen” OR “frail elder” OR “very old” OR ag*ing OR “in old age”)
#2	AB (Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “web-based” OR “Internet-based” OR “Internet intervention” OR ICT OR “communications media” OR “information technology”) OR TI (computer* OR Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “web-based” OR “Internet-based” OR “Internet intervention” OR ICT OR “communications media” OR “information technology”)
	AB (Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “web-based” OR “Internet-based” OR “Internet intervention” OR ICT OR “communications media” OR “information technology”) OR TI (computer* OR Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “web-based” OR “Internet-based” OR “Internet intervention” OR ICT OR “communications media” OR “information technology”)
#3	S1 AND S2
CINAHL
#1	TI (computer* OR Internet* OR “digital divide” OR technogenerian* OR “silver surfer” OR “information society” OR “e-mail” OR chat OR “e-resource” OR “e-intervention” OR “Internet-mediated intervention” OR online* OR “Internet intervention” OR ICT OR “communications media” or “social media”) OR MJ (“Internet” or “world wide web” or “communications media” or “social networks” or “attitude to computers”)
#2	TI (elder* OR geriatr* OR “older adult” OR “older people” OR “late* life” OR “late adulthood” OR senil* OR “senior citizen” OR “frail elder” OR “very old” OR aging OR residential OR “in old age”) OR MJ (“Aged+” or “Aging+”)
#3	S1 AND S2

Acknowledgements

The authors thank Information Specialist Pia Pörtfors (THL) for valuable help with finalizing literature database search strategies, and Dr Carolina Herberts for thorough language review. They also thank the reviewers of the manuscript for valuable suggestions and comments during the work process.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors have received funding for a small research project which this current study is a part of. The funding bodies are Svenska Kulturfonden, Signe och Ane Gyllenbergs Stiftelse, and Svensk-Österbottniska Samfundet. The funding bodies support research projects in Finland, they do not focus on specific research topics/issues and present no conflict of interest as they have not at any point been involved in the research process.

Author Biographies

Anna K. Forsman is a Senior Researcher specialized in mental health in older people and psychosocial covariates. She holds a position as a University Teacher in Developmental Psychology at Åbo Akademi University in Vaasa, Finland.

Johanna Nordmyr is a Doctoral Student in Developmental Psychology at Åbo Akademi University in Vaasa, Finland. Her research interests include Internet use and various mental health factors across the lifespan.

References

Aguilar

Boerema

Harris

(2010). Meanings attributed by older adults to computer use. Journal of Occupational Science, 17, 27-33.

Blazun

Saranto

Rissanen

(2012). Impact of computer training courses on reduction of loneliness of older people in Finland and Slovenia. Computers in Human Behavior, 28, 1202-1212.

Bohlmeijer

Smit

Cuijpers

(2003). Effects of reminiscence and life-review on late-life depression: Meta-analysis. International Journal of Geriatric Psychiatry, 18, 1088-1094.

Bowling

A. P.

(2007). Aspirations for older age in the 21st century: What is successful ageing? International Journal of Ageing and Human Development, 64, 263-297.

Bronfenbrenner

(1979). The ecology of human development. Cambridge, MA: Harvard University Press.

Bryant

Corbett

Kutner

(2001). In their own words: A model of healthy aging. Social Science & Medicine, 53, 927-941.

Burbank

P. M.

(1986). Psychosocial theories of aging: A critical evaluation. Advances in Nursing Science, 9, 73-86.

Carey

Elton

M. C. J.

(2010). When media are new: Understanding the dynamics of new media adoption and use. Ann Arbor: University of Michigan Press.

Cutler

S. J.

(2005). Ageism and technology. Generations, 29(3), 67-72.

10.

European Commission. (2013). European social statistics 2013 edition. Retrieved from http://ec.europa.eu/eurostat/documents/3930297/5968986/KS-FP-13-001-EN.PDF/

11.

Forsman

A. K.

(2012). The importance of social capital in later life: Mental health promotion and mental disorder prevention among older adults (Doctoral thesis). Nordic School of Public Health NHV, Gothenburg, Sweden.

12.

Forsman

A. K.

Nordmyr

Wahlbeck

(2011). Psychosocial interventions for the promotion of mental health and the prevention of depression among older adults. Health Promotion International, 26(Suppl. 1), i85-i107.

13.

Forsman

A. K.

Ventus

D. B. J.

van der Feltz-Cornelis

C. M.

Wahlbeck

(2014). Public mental health research in Europe: A systematic mapping for the ROAMER project. European Journal of Public Health, 24, 955-960.

14.

FUTURAGE. (2011). Road map for European ageing research. European Commission. Retrieved from http://www.futurage.group.shef.ac.uk/road-map.html

15.

Greenfield

E. A.

(2011). Using ecological frameworks to advance a field of research, practice, and policy on aging-in-place initiatives. The Gerontologist, 52, 1-12.

16.

Hannes

(2011). Critical appraisal of qualitative research. In Noyes

Booth

Hannes

Harden

Harris

Lewin

Lockwood

(Eds.), Supplementary guidance for inclusion of qualitative research in Cochrane systematic reviews of interventions. Cochrane Collaboration Qualitative Methods Group. Retrieved from http://cqrmg.cochrane.org/supplemental-handbook-guidance

17.

Harden

(2010). Mixed-methods systematic reviews: Integrating quantitative and qualitative findings (Technical Brief No.25). National Center for the Dissemination of Disability Research. Retrieved from http://www.ncddr.org/kt/products/focus/focus25/

18.

Havighurst

R. J.

Albrecht

(1953). Older people. New York, NY: Longmans, Green.

19.

Hesse-Biber

S.N.

(2010). Mixed Methods Research: Merging Theory with Practice. New York: The Guilford Press.

20.

Knapp

McDaid

Parsonage

(Eds.). (2011). Mental health promotion and mental illness prevention: The economic case. London, England: Department of Health.

21.

Krippendorff

(2004). Content analysis: An introduction to its methodology. Thousand Oaks, CA: Sage.

22.

Lahtinen

Lehtinen

Riikonen

Ahonen

(Eds.). (1999). Framework for promoting mental health in Europe. Hamina, Finland: Ministry of Social Affairs and Health, STAKES, National Research and Development Centre for Welfare and Health.

23.

Lemon

B. W.

Bengtson

V. L.

Peterson

J. A.

(1972). Exploration of the activity theory of aging: Activity types and life satisfaction among in-movers to a retirement community. Journal of Gerontology, 27, 511-523.

24.

Moher

Liberati

Tetzlaff

Altman

D. G.

, & the PRISMA Group. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine, 6(6): Article e1000097. doi:10.1371/journal.pmed1000097

25.

Noyes

Lewin

(2011a). Extracting qualitative evidence. In Noyes

Booth

Hannes

Harden

Harris

Lewin

Lockwood

(Eds.), Supplementary guidance for inclusion of qualitative research in Cochrane systematic reviews of interventions (Version 1). Cochrane Collaboration Qualitative Methods Group. Retrieved from http://cqrmg.cochrane.org/supplemental-handbook-guidance

26.

Noyes

Lewin

(2011b). Supplemental guidance on selecting a method of qualitative evidence synthesis, and integrating qualitative evidence with Cochrane intervention reviews. In Noyes

Booth

Hannes

Harden

Harris

Lewin

Lockwood

27.

Patton

M. Q.

(2002). Qualitative research & evaluation methods (3rd ed.). Thousand Oaks, CA: Sage.

28.

Pearson

(2004). Balancing the evidence: Incorporating the synthesis of qualitative data into systematic reviews. The Joanna Briggs Institute Report, 2(2), 45-64.

29.

Pew Research Center. (2013). Demographics of internet users. Pew Internet & American Life Project. Retrieved from http://pewinternet.org/Static-Pages/Trend-Data-(Adults)/Whos-Online.aspx

30.

Seybert

(2012). Internet use in households and by individuals in 2012 (Statistics in focus, 50/2012). Eurostat, European Commission. Retrieved from http://ec.europa.eu/eurostat/documents/3433488/5585460/KS-SF-12-050-EN.PDF/39000dab-e2b7-49b2-bc4b-6aad0bf01279

31.

Shapira

Barak

Gal

(2007). Promoting older adults’ well-being through Internet training and use. Aging & Mental Health, 11, 477-484.

32.

United Nations, Department of Economic and Social Affairs, Population Division. (2002). World population ageing: 1950-2050. New York, NY: United Nations Publishing. Retrieved from http://www.un.org/esa/population/publications/worldageing19502050/pdf/preface_web.pdf

33.

van der Wardt

Bandelow

Hogervorst

(2012). The relationship between cognitive abilities, wellbeing and use of new technologies in older people. Gerontechnology, 10(4), 187-207.

34.

World Health Organization. (1986). Ottawa charter for health promotion. Ottawa, Ontario, Canada: Author. Retrieved from http://www.who.int/healthpromotion/conferences/previous/ottawa/en/

35.

World Health Organization. (2005). Promoting mental health: Concepts, emerging evidence, practice. Geneva, Switzerland: Author. Retrieved from http://www.who.int/mental_health/evidence/MH_Promotion_Book.pdf

36.

World Health Organization. (2013). What is “active ageing”? Author. Retrieved from http://www.who.int/ageing/active_ageing/en/

37.

World Health Organization. (2014). Mental health. Author. Retrieved from http://www.who.int/topics/mental_health/en/

38.

Berner

Rennemark

Jogreus

Berglund

(2012). Distribution of personality, individual characteristics and internet usage in Swedish older adults. Aging and Mental Health, 16, 119-126.

39.

Carpenter

B. D.

Buday

(2007). Computer use among older adults in a naturally occurring retirement community. Computers in Human Behavior, 23, 3012-3024.

40.

Choi

N. G.

DiNitto

D. M.

(2013a). Internet use among older adults: Association with health needs, psychological capital, and social capital. Journal of Medical Internet Research, 15(5), e97.

41.

Choi

N. G.

DiNitto

D. M.

(2013b). The digital divide among low-income homebound older adults: Internet use patterns, ehealth literacy, and attitudes toward computer/internet use. Journal of Medical Internet Research, 15(5), e93.

42.

Clark

Jackson

Carlson

Chou

C. P.

Cherry

Jordan-Marsh

Azen

S. P.

(2011). Effectiveness of a lifestyle intervention in promoting the wellbeing of independently living older people: Results of the Well Elderly 2 randomized controlled trial. Journal of Epidemiology and Community Health, 66, 782-790.

43.

Cotten

S. R.

Ford

Hale

T. M.

(2012). Internet use and depression among older adults. Computers in Human Behavior, 28, 496-499.

44.

Cresci

M. K.

Yarandi

H. N.

Morrell

R. W.

(2010). The digital divide and urban older adults. Computers, Informatics, Nursing, 28(2), 88-94.

45.

Erickson

Johnson

G. M.

(2011). Internet use and psychological wellness during late adulthood. Canadian Journal on Aging, 30, 197-209.

46.

Heo

Kim

Won

(2011). Exploring the relationship between internet use and leisure satisfaction among older adults. Activities, Adaptation & Aging, 35, 43-54.

47.

Jackson

Mandel

Blanchard

Carlson

Cherry

Azen

Clark

F. A.

(2009). Confronting challenges in intervention research with ethnically diverse older adults: The USC Well Elderly II trial. Clinical Trials, 6, 90-101.

48.

James

B. D.

Boyle

P. A.

Bennett

D. A.

(2013). Internet use and decision making in community-based older adults. Frontiers in Psychology, 4, Article 605.

49.

Koopman-Boyden

Reid

S. L.

(2009). Internet/e-mail usage and well-being among 65-84 year olds in New Zealand: Policy implications. Educational Gerontology, 35, 990-1007.

50.

Lagergren

Fratiglioni

Rahm Hallberg

Berglund

Elmståhl

Hagberg

. . . Wimo

(2004). A longitudinal study integrating population, care and social services data: The Swedish national study on aging and care (SNAC). Aging Clinical and Experimental Research, 16, 158-168.

51.

Lam

(2009). The use of information technology and mental health among older care-givers in Australia. Aging and Mental Health, 13, 557-562.

52.

Lelkes

(2013). Happier and less isolated: Internet use in old age. Journal of Poverty and Social Justice, 21, 33-46.

53.

Medeiros

F. D. L.

Xavier

A. J.

Schneider

I. J.

Ramos

L. R.

Sigulem

d’Orsi

(2012). Digital inclusion and functional capacity of older adults living in Florianopolis, Santa Catarina, Brazil (EpiFloripa 2009-2010). Revista Brasileira de Epidemiologia (Brazilian Journal of Epidemiology), 15(4), 106-122.

54.

Näsi

Räsänen

Sarpila

(2012). ICT activity in later life: Internet use and leisure activities amongst senior citizens in Finland. European Journal of Ageing, 9, 169-176.

55.

Stark-Wroblewski

Edelbaum

J. K.

Ryan

J. J.

(2007). Senior citizens who use e-mail. Educational Gerontology, 33(4), 293-307.

56.

Sum

Mathews

R. M.

Hughes

Campbell

(2008). Internet use and loneliness in older adults. Cyberpsychology and Behavior, 11, 208-211.

57.

Sum

Mathews

M. R.

Pourghasem

Hughes

(2008). Internet technology and social capital: How the Internet affects seniors’ social capital and wellbeing. Journal of Computer-Mediated Communication, 14, 202-220.

58.

Sum

Mathews

R. M.

Pourghasem

Hughes

(2009). Internet use as a predictor of sense of community in older people. Cyberpsychology and Behavior, 12, 235-239.

59.

Werner

J. M.

Carlson

Jordan-Marsh

Clark

(2011). Predictors of computer use in community-dwelling, ethnically diverse older adults. Human Factors, 53, 431-447.

60.

Zhang

Umemuro

(2012). Older adults’ hedonic and eudaimonic well-being in information and communication technology activities. Gerontechnology, 10(4), 231-243.

61.

Aguilar

Boerema

Harris

(2010). Meanings attributed by older adults to computer use. Journal of Occupational Science, 17, 27-33.

62.

Beckenhauer

J. I. L.

Armstrong

(2009). Exploring relationships between normative aging, technology, and communication. Marriage & Family Review, 45, 825-844.

63.

Cresci

M. K.

Novak

J. M.

(2012). Information technologies as health management tools: Urban elders’ interest and ability in using the internet. Educational Gerontology, 38, 491-506.

64.

Gatto

S. L.

Tak

S. H.

(2008). Computer, internet, and e-mail use among older adults: Benefits and barriers. Educational Gerontology, 34, 800-811.

65.

Khvorostianov

Elias

Nimrod

(2012). “Without it I am nothing”: The internet in the lives of older immigrants. New Media & Society, 14, 583-599.

66.

Larsson

Larsson-Lund

Nilsson

(2013). Internet based activities (IBAs): Seniors’ experiences of the conditions required for the performance of and the influence of these conditions on their own participation in society. Educational Gerontology, 39(3), 155-167.

67.

Malta

(2007). Love actually! Older adults and their romantic internet relationships. Australian Journal of Emerging Technologies and Society, 5(2), 84-102.

68.

Pfeil

Zaphiris

Wilson

(2009). Older adults’ perceptions and experiences of online social support. Interacting With Computers, 21(3), 159-172.

69.

Russell

Campbell

Hughes

(2008). Ageing, social capital and the internet: Findings from an exploratory study of Australian “silver surfers.” Australasian Journal on Ageing, 27(2), 78-82.

70.

Russell

(2011). Later life ICT learners ageing well. International Journal of Ageing & Later Life, 6(2), 103-127.

71.

Sayago

Blat

(2010). Telling the story of older people e-mailing: An ethnographical study. International Journal of Human-Computer Studies, 68, 105-120.

72.

Sourbati

(2009). “It could be useful, but not for me at the moment”: Older people, internet access and e-public service provision. New Media Society, 11, 1083-1100.

73.

Whitbourne

S. K.

Ellenberg

Akimoto

(2013). Reasons for playing casual video games and perceived benefits among adults 18 to 80 years old. Cyberpsychology, Behavior, and Social Networking, 16, 892-897.

74.

Xie

(2007). Using the internet for offline relationship formation. Social Science Computer Review, 25, 396-404.