Morrow-Howell N, Hinterlong J, Rozario PA, Tang F. Effects of volunteering on the well-being of older adults. J Gerontol B Psychol Sci Soc Sci. 2003; 58(3):S137-45.
Introduction:
This study examined the effects of volunteering on the well-being of older adults, using data from the longitudinal Americans Changing Lives Study (ACL) – which had three waves. It found that volunteer status and number of volunteer hours was significantly linked to improved well-being, and fewer depressive symptoms, specifically, in older adults. The researchers found that the maximum benefit from volunteering was achieved when older adults volunteer 100 hours a year, or as little as 2 hours per week. The number of organizations or types of organizations older adults volunteered with did not influence well-being.
Construct Validity:
Several different researchers have analyzed data from the ACL – but they used more crude measures of volunteering (e.g. volunteering or not) and well-being/happiness (e.g. depressive symptomatology). In this analysis, the researchers looked at many different facets of volunteering: volunteering status, number of hours spent volunteering, type of volunteer organization, number of organizations the participant volunteered with, perceived benefit of volunteer work.
The researchers defined well-being as (lack-of) depressive symptomology, self-rated health, and self-reported functional dependency.
External Validity:
The ACL study is a national survey of adults that oversampled African-Americans and adults over 60. This study was designed to yield results that may be generalized to adults in the U.S. over the age of 60.
Internal Validity:
A Cohort Study
Prospective vs. Retrospective
Participants/Recruitment:
Data was obtained from the Americans Changing Lives Study (ACL), a three-wave longitudinal survey of 3617 American adults. The authors of this study used a subset of individuals 60 years of age and older. In wave 1 (1986) 1669 respondents were included in the analysis, and at Wave 2 (1989), 1,279 older respondents were reinterviewed. At Wave 3 (1994), over 900 of these respondents were located and reinterviewed. The authors used multiple random imputation to account for non-deceased respondents who dropped out in subsequent waves of the study.
Inclusion Criteria:
- Age 60 and above.
- No other criteria listed.
Exclusion Criteria:
- Not listed.
This study was longitudinal in design, and is vulnerable to what the researchers call “social selection” a type of selection bias which can arise if those who refuse to be in the study or who are lost to follow-up differ from those who enter the study with respect to the variables of interest (i.e. volunteering and well-being). In this study, it is worrisome that nearly 30% of those who entered the study had deceased by the third wave.
The researchers controlled for the following potential confounders: demographic characteristics, informal social integration, and well-being at the previous study wave.
A strength of this study is that it controlled for a range of potential confounding variables, including self-rated health, depression and functional disability at the previous time point, in order to prospectively look at the association between volunteering and wellbeing.
Conclusion:
The results of this study suggested a nonlinear relationship between hours of volunteering and well-being outcomes. Volunteers who were involved for approximately 100 hours a year (2-3 hours a week) were found to have the highest measures of well-being. Higher levels of involvement were not associated with increasing gains. These results suggest that even relatively low-levels of volunteering can lead to positive outcomes.
Like previous work on the effects of volunteering, the authors of the study faced the difficult problem of social causation versus social selection. As this study was a retrospective analysis, it provides little information as to whether the increase in well-being was due to volunteering or whether those with better well-being are more likely to volunteer. The authors of this study posited that a combination of both may be responsible for the association.
Yuen HK, Huan P, Burik JK, Smith TG. Impact of participating in volunteer activities for residents living in long-term-care facilities. The American Journal of Occupational Therapy. 2008; 62 (1):71-76.
Introduction:
This study is a randomized control trial, where 49 residents at 5 different long-term care facilities, all over the age of 60, were randomly assigned to one of two groups: the experimental group (who volunteered with an English-as-a-second-language program for 3 months) and a control group (who received usual care at their nursing home). The effects of the volunteer program on well-being (life satisfaction, depression, and self-rated health) was measured at 3 months (at the end of the program) and at 6 months (3 months after the study ended). The researchers found that at the end of the program, residents who volunteered had overall higher well-being in comparison to residents who did not volunteer. Three months after the program ended, these effects persisted. Although this study was small in scale, used a very specific form of volunteering and involved a unique population, it is one of the only random control trials examining the effects of volunteering on well-being (or depression or life satisfaction).
Construct Validity:
The volunteer activity residents were randomized into two groups. In the intervention group, residents volunteered (twice a week for 12 weeks) to teach English conversation skills to English-as-a-second-language (ESL) students.
The researchers measured well-being as an overall composite of depression, self-rated health, and life satisfaction. Depression was measured using the Geriatric Depression Scale. Life satisfaction was measured with a validated scale known as the Life Satisfaction Index-A. Self-rated health was measured with one question, asking residents to rate their health on a scale from 1 (poor) to 5 (excellent).
External Validity:
The results of this study can be properly generalized to elderly residents of long-term care facilities, and possibly the elderly population at large. The volunteering program used in this study involved one-on-one meetings twice a week between the resident and a foreign student. Less structured volunteering and less social forms of volunteering may not necessarily produce the same effects.
Internal Validity:
Prospective vs. Retrospective
Open trial vs. Single blinded vs. Double blinded
The research assistants administering the questionnaires to the residents were unaware of the purpose of the study, which is a strong-point, as it reduces the chance that the researcher’s expectations affected how participants responded.
Nevertheless, studies that investigate the effect of an intervention that requires active participation (such as volunteering) cannot have a true “placebo” control, and the participants cannot be blinded. Due to the active nature of the intervention, participants were aware of whether they were part of the treatment or control group. The researchers therefore could not control for the “Hawthorne effect,” that is that subjects may change their behavior simply because they are being studied.
Randomized vs. Nonrandomized
Participants were randomized into two groups, a “usual care” control group, and a volunteer intervention group. Study subjects were additionally stratified by mental function and long-term care facility.
Participants/Recruitment:
All participants were recruited from a group of residents of five long-term care facilities (1 skilled nursing, 3 assisted living, and 1 semi-independent living) in Charleston County, South Carolina.
Inclusion Criteria:
- 60 year of age or older.
- Resident of long-term care facility.
- Native English speaker.
- Be able to carry on a conversation appropriately for at least 1 hr.
- Mini-Mental State Examination score of 19 or greater.
Exclusion Criteria:
- Uncorrectable hearing impediment.
- Maladaptive behavioral pattern or exhibition of overt psychotic symptoms.
- Terminal illness with a life expectancy of less than 6 months.
The study design was a randomized controlled trial, considered the “gold standard” in study designs. Randomizing study participants to either the usual care or intervention group allows for each treatment group to be as equivalent as possible with respect to known and unknown factors that could confound the association between variables. However, this study was fairly small and only 39 participants were randomized to each group. This small sample size diminished the effect of randomization, as the authors could not be certain that the two groups were adequately stratified. To help ensure that each of the two groups were as equal as possible, the researchers stratified the groups according to mental function and individual long-term care facility.
The drop-out rate for the study, including those lost to follow-up, was 30%. More residents were lost to follow-up in the control “usual care” group. It is possible these residents were not a representative subset, thus biasing the results of the study.
Conclusion:
The researchers of this study found that at the end of the intervention, residents in the volunteering group showed overall higher well-being than residents in the usual care group. This is a reasonable conclusion based on the results. The researchers also found that three months after the study ended, these positive effects persisted. When the researchers looked at depression, life satisfaction, and health separately, the findings were not significant; this, however, may have been due to small sample-size.
Musick MA, Wilson J. Volunteering and depression: The role of psychological and social resources in different age groups. Social Science and Medicine. 2003; 56:259-269.
Introduction:
This study is a retrospective analysis of data from the Americans’ Changing Lives (ACL) study, a large cross-sectional survey of persons aged 25 and over living in the United States. The authors sought not only to elucidate relationship between volunteering and depression, but also to identify the role of social and psychological resources in mediating the relationship. The study found that volunteering and depression are inversely related in those over the age of 65. Interestingly, the authors of the study found evidence to support the notion that some of the effect of volunteering on depression may be mediated by the social integration in encourages.
Construct Validity:
Depression was measured using an 11-item version of the Center for Epidemiologic Studies Depression scale (CES-D). Volunteering was assessed at three time points, and divided into five categories: religious volunteering, school/educational volunteering, political group/labor union volunteering, senior citizen group volunteering, any other volunteering. A variable named “sustained volunteering” was constructed by using information about whether individuals were volunteering at each of the three surveyed time points. “Social resources” were measured by asking respondents how often they attended meetings of groups or clubs, and by asking how often respondents got together (either on the phone or in person) with friends and relatives. “Psychological resources” were measured using a three-item index of self-esteem.
External Validity:
The Americans’ Changing Lives study is a geographically diverse survey of residents of the contiguous 48 states. The study, however, sampled African-Americans and persons aged 60 and over at twice the rate of non-African-Americans and those under the age of 60.
Internal Validity:
A Retrospective Cohort Study
Prospective vs. Retrospective
The following potential confounders were measured at each assessment and accounted for via adjusted analysis:
Health Conditions (functional impairment, and number of life threatening conditions)
Physical Activity (Likert scale ranging from 1-never to 4-often, indicating how often people worked in their yard, exercised/played sports, and took walks)
Race
Employment (in labor force or not)
Church Attendance (Likert scale ranging from 1-never, to 6-more than once a week)
Married or not
Income
Age (years)
Gender
Participants/Recruitment:
3617 participants were surveyed in the first wave of the American’s Changing Lives study. Subjects were surveyed at three points in time: 1986, 1989 and 1994. By the third wave, 35% of the original participants had dropped out, including 19% due to death. Inclusion/Exclusion criteria for the ACL study were not listed.
As a retrospective study, associations between variables were analyzed, and no cause-effect relationships could be ascertained.
Conclusion:
The researchers concluded that volunteering is associated with lower levels of depression in those over 65, and those who engaged in “sustained volunteering” tended to be less depressed no matter what their age.
Volunteering at all three time points was associated with a trend towards decreased depression in younger adults. This association, however, was not statistically significant. “Social resources” were found to be correlated with levels of both volunteering and depression, providing support for the hypothesis that increased social integration mediates some of the effect of volunteering on depression.
Hao Y. Productive Activities and Psychological Well-Being Among Older Adults. J Gerontol B Psychol Sci Soc Sci. 2008; 63(2):S64-72
Introduction:
The author of this study sought to test whether paid work and/or formal volunteering reduces the rate of mental health decline in later life. Data was obtained from the Health and Retirement Study (HRS), a longitudinal survey of nationally representative individuals from the 1931-1941 birth cohorts. The study analyzed HRS data obtained from 1996-2002. Surveyed participants who were active in paid work and volunteering were found to have better mental health at baseline. Moreover, those who continued to be employed full-time and volunteered on a part-time basis were found to have a slower rate of mental health decline.
Construct Validity:
Volunteering was divided into low level volunteering (less than 100 hrs/year) and high-level volunteering (100 hours/yr or more).
Paid work was divided into part-time work (less than 1680 hrs/year) and full-time work (1680 hrs/year or more).
“Psychological Well-Being” was assessed using the eight-item Center for Epidemiologic Studies-Depression (CES-D) scale.
External Validity:
The sample included between 5,000 and 7,000 (due to attrition) older adults recruited from the Health and Retirement Study (HRS) between the ages of 55 and 66. Subjects were followed over a 6 year period. Approximately 15% were Black, 8% Hispanic and the remainder were white.
Internal Validity:
A Cohort Study
Prospective vs. Retrospective
The following potential confounders were measured at each assessment and accounted for via adjusted analysis:
Self-rated health (score from 1 to 5)
Functional disability (score from 1 to 80)
Index of chronic conditions (sum of number of conditions)
Marital Status (measured at each time point)
Race (black, Hispanic, or white)
Income (continuous in $10,000)
Age (years)
Gender
Participants/Recruitment:
Data was obtained from the Health and Retirement Study (HRS), a longitudinal survey of nationally representative individuals from the 1931-1941 birth cohorts. At baseline, a total of 7830 respondents were recruited. Subjects were followed over a 6 year period.
Inclusion Criteria:
-Age between 55 and 66.
-No other criteria listed.
Exclusion Criteria:
-Not listed.
Conclusion:
The researcher found that any volunteering was associated with better mental health at baseline. In addition, he found that low levels of volunteering in conjunction with paid work were associated with slower rates of mental decline at later points in time.
Volunteering alone did not predict slower rates of mental decline. The author of the study suggests two possible explanations for this: 1) participants may have substituted paid work with volunteering. 2) selection bias, i.e. those who had better mental health were more likely to volunteer to begin with, and thus further volunteering did not show a mental health benefit.