Abstract
Background
The association between hobby engagement and risk of dementia reported from a short-term follow-up study for individuals aged ≥65 years may be susceptible to reverse causation. We examined the association between hobby engagement in age of 40–69 years and risk of dementia in a long-term follow-up study among Japanese, including individuals in mid-life, when the majority of individuals have normal cognitive function.
Methods
A total of 22,377 individuals aged 40–69 years completed a self-administered questionnaire in 1993–1994. The participants answered whether they had hobbies according to the three following responses: having no hobbies, having a hobby, and having many hobbies. Follow-up for incident disabling dementia was conducted with long-term care insurance data from 2006 to 2016.
Results
During a median of 11.0 years of follow-up, 3,095 participants developed disabling dementia. Adjusting for the demographic, behavioral, and psychosocial factors, the multivariable hazard ratios of incident disabling dementia compared with “having no hobbies” were 0.82 (95% confidence interval [CI], 0.75–0.89) for “having a hobby” and 0.78 (95% CI, 0.67–0.91) for “having many hobbies”. The inverse association was similarly observed in both middle (40–64 years) and older ages (65–69 years). For disabling dementia subtypes, hobby engagement was inversely associated with the risk of dementia without a history of stroke (probably non-vascular type dementia), but not with that of post-stroke dementia (probably vascular type dementia).
Conclusion
Hobby engagement in both mid-life and late life was associated with a lower risk of disabling dementia without a history of stroke.
Key words: hobby engagement, disabling dementia, follow-up study, epidemiology
INTRODUCTION
A hobby is defined as an enjoyable leisure activity in which individuals actively engage at their own initiative in their free time from work or other responsibilities.1 A hobby includes cognitive and physical leisure activities, and engagement in these activities was associated with a lower risk of dementia.2–13 Enjoyable leisure activities were also correlated with a higher level of life engagement (purpose in life),1 probably lowering dementia risk.14 However, evidence on the relationship between hobby engagement and the risk of dementia is limited.6,8 Furthermore, previous studies examining the association between hobby engagement and risk of dementia recruited individuals aged ≥65 years and followed up their participants for less than 7 years at mean,6,8 probably resulting in reverse causation because a cognitive decline may restrict engagement in hobbies. To minimize the impact of reverse causation, it is necessary to evaluate hobby engagement in mid-life, when the majority of individuals have normal cognitive function, and/or conduct a long-term follow-up survey.
Therefore, we aimed to examine the association between hobby engagement in age of 40–69 years and the long-term risk of dementia in the Japan Public Health Center-based prospective (JPHC) Study Cohort II. We hypothesized that having hobbies was associated with a lower risk of dementia.
METHODS
Study population
The JPHC Study Cohort II was a population-based cohort study that started in 1993, enrolling residents aged 40–69 years from six PHC areas: Nagaoka, Niigata Prefecture; Mito, Ibaraki Prefecture; Suita, Osaka Prefecture; Chuo-higashi, Kochi Prefecture; Kamigoto, Nagasaki Prefecture; and Miyako, Okinawa Prefecture. A detailed description of the JPHC study protocol has been published elsewhere.15 Briefly, the participants were provided with a self-administered questionnaire about demographic characteristics, medical history, physical activity, smoking, drinking, socioeconomic status, and dietary habits in 1993–1994, and thereafter, were followed up on their health outcomes of disabling dementia ascertained from the long-term care insurance (LTCI) data. Under the Japanese LTCI system starting in 2000, the long-term care service was provided with co-payment to persons aged ≥65 years who need support or care for their activities of daily living (ADL), and those aged 40–64 years who need care due to having specified aging-related diseases, including dementia.16,17 In four towns (Iwase district in Sakuragawa city and Tomobe district in Kasama city in Ibaraki Prefecture; and Kagami and Noichi districts in Konan city in Kochi Prefecture), the LTCI data were available from January 1, 2006, to December 31, 2016, through the approval of local municipalities for the provision of the LTCI data. A total of 13,251 men and 13,976 women were included in the study.
In the current study, we excluded participants with ineligible criteria: non-Japanese nationality, late report of migration occurring before the self-administered questionnaire survey, duplicate registration, or refusal to follow-up survey (n = 26); and those who died, moved out of the study area, or were lost to follow-up before January 1, 2006 (n = 4,129). Participants who had a history of stroke in the self-administered questionnaire survey in 1993–1994 (n = 146) and/or missing information about their hobbies (n = 549) were also excluded. Finally, a total of 10,405 men and 11,972 women were included in the current analysis (Figure 1).
Figure 1. Flowchart of study participants.
The participants were informed of the objectives of the study, and the completion of the survey questionnaire in 1993–1994 was regarded as providing consent for participation. This study was approved by the Institutional Review Boards of the National Cancer Center, Japan, and Osaka University.
The self-administered questionnaire survey
The participants were queried about their hobby engagement as follows: “Do you have any hobbies?”, with three possible responses: having no hobbies, having a hobby, and having many hobbies. Body mass index was calculated as the weight (kg) divided by height squared (m2). We calculated the total physical activity by summing all the products of daily engagement time and metabolic equivalents (METs) per hour in METs/day for each activity (strenuous exercise, sitting, standing or walking, and sleep or others).18 Our previous study using the 5-year follow-up questionnaire of the JPHC study, which included the same items of physical activity in this study, showed that the Spearman’s rank correlation coefficient between the total METs/day score and physical activity record was 0.46.18 We also asked the participants, “How many friends do you meet at least once per week?”, with three possible responses: no friends, 1–3 friends, and ≥4 friends. Type A characteristics were assessed based on four items representing the aspects of competitive drive, speed and impatience, aggressiveness, and irritability. We calculated the type A behavior pattern index by summing the scores of four items (0 to 2 for each) and divided the participants into four categories: low (score of 0 to 3), medium (score of 4), high (score of 5), and very high (score of 6 or more).19 The participants answered about perceived mental stress from “How much stress do you have in your daily life?”, with three responses: low, moderate, and high. We further asked for the participant, “Are you living with someone?”, with any of five possible responses: alone, spouse, child(ren), parent(s), and others. The status of hypertension, diabetes, and hypercholesterolemia were determined from the responses to the baseline questions regarding the use of medication and the medical history of the respective diseases.
Definition of disabling dementia
In the Japanese LTCI system, when the certification of care level was applied, the preliminary care level of applicants was computed based on a structured interview survey. Subsequently, the committee of long-term care requirement certification in the local government, which is composed of experts in medical, health, and welfare areas, finalized the care level for the applicant after reviewing all the documents, including the result of the computerized preliminary care level and the statement of the primary doctor’s examination.16,17 We defined disabling dementia as individuals who were certified regarding their care level (excluding support levels) and ranked in IIa or worse grade of ADL in older adults with dementia by the attending physician. The first certified date of the care level meeting with our dementia criteria was used as the date of the incident disabling dementia. These criteria were previously verified in comparison with neuropsychiatrists’ diagnoses of disabling dementia, and the sensitivity and specificity were 73% and 96%, respectively.20
To investigate whether the association between hobby engagement and disabling dementia differed with dementia subtypes, we divided disabling dementia cases into disabling dementia without a history of stroke (possibly non-vascular type dementia) and post-stroke disabling dementia (possibly vascular-type dementia) according to the presence or absence of a history of stroke before the onset of disabling dementia. A history of stroke was obtained from 5- and 10-year follow-up questionnaires and stroke registration. The details of stroke registration have been published elsewhere.21
Statistical analysis
Person-years were calculated for each individual as the duration from January 1, 2006, when the outcome data started to be available, to the date of identification of disabling dementia, the date of death, lost to follow-up, or the end of follow-up (December 31, 2016), whichever occurred first. When we used the dementia subtypes as the outcomes, the follow-up ended on December 31, 2012, instead of December 31, 2016, owing to the unavailability of the stroke registry data.
Differences in mean value or proportions of the baseline characteristics were tested using the chi-square test and the analysis of variance. The hazard ratios (HRs) and 95% confidence intervals (CIs) of incident disabling dementia were calculated according to the hobby engagement categories using the Cox proportional hazards model. The proportional hazard assumption was confirmed visually by looking at the log-log plots using LIFETEST procedure and was not rejected by assessing the weighted Schoenfeld residuals using PHREG procedures with zph-options. We tested the interaction of hobby engagement with other variables for incident disabling dementia using cross-product terms in the Cox proportional hazards model. As there was no interaction between having hobbies and sex in relation to the incident disabling dementia (P for interaction: 0.28 for having a hobby and 0.60 for having many hobbies), we did not conduct sex-specific analyses. The HRs and 95% CIs of the disabling dementia subtypes were calculated using a cause-specific Cox proportional hazard model. In model 1, we stratified jointly by area (four towns) and adjusted for age (years) at the questionnaire survey in 1993–1994 and sex. In model 2, we adjusted further for body mass index (<18.5, 18.5–<25.0, 25.0–<30, ≥30 kg/m2, and missing), smoking status (never, former, 1–19 cigarettes/day, ≥20 cigarettes/day, and missing), alcohol intake (non-drinkers, former drinker, occasional drinkers, 1–<150 g/week, 150–<300 g/week, ≥300 g/week, and missing), total physical activity (tertiles of METs/day, and missing), and histories of hypertension, diabetes, and hypercholesterolemia (yes, no, and missing, for each). In model 3, we adjusted further for living alone (yes, no, and missing), job status (yes, no, and missing), perceived mental stress (low, moderate, high, and missing), type A characteristics (low, medium, high, very high, and missing), and the number of friends (no friends, 1–3 friends, ≥4 friends, and missing). We used the category of missing data for these confounding variables in the statistical model. For sensitivity analysis, we excluded the participants with missing data. We also conducted stratified analysis by age categories at the questionnaire survey in 1993–1994 (40–64 years and 65–69 years old).
All statistical analyses were performed using the SAS (version 9.4; SAS Institute, Cary, NC, USA) in a two-sided test. The statistical significance was set at P < 0.05.
RESULTS
Table 1 shows the baseline characteristics of potential confounding factors according to the hobby engagement categories. Individuals having hobbies were younger, had higher proportions of men, being physically active, current smoker, current drinker, medical history of hypercholesterolemia, being employed, living alone, having low mental stress, high type A characteristics, and having friends, and lower proportions of medical history of hypertension and medication use for hypertension compared with those having no hobbies.
Table 1. Characteristics of participants in 1993–1994 according to the hobby engagement categories.
| Hobby categories |
P for difference |
|||||||
|
| ||||||||
| Having no hobbies | Having a hobby | Having many hobbies | ||||||
| Number at risk | 4,518 | 16,126 | 1,733 | |||||
| Age, years, mean (SD) | 53.1 | (8.8) | 52.5 | (8.5) | 52.7 | (8.3) | <0.001 | |
| Sex, n (%) | ||||||||
| Men | 1,692 | (37.5) | 7,765 | (48.2) | 948 | (54.7) | <0.001 | |
| Women | 2,826 | (62.6) | 8,361 | (51.9) | 785 | (45.3) | ||
| Body mass index | ||||||||
| kg/m2, mean (SD) | 23.3 | (3.1) | 23.3 | (2.9) | 23.4 | (2.9) | 0.28 | |
| n (%) | <18.5 kg/m2 | 198 | (4.4) | 524 | (3.3) | 43 | (2.5) | |
| 18.5–<25.0 kg/m2 | 3,098 | (68.6) | 11,410 | (70.8) | 1,235 | (71.3) | ||
| 25.0–<30.0 kg/m2 | 1,024 | (22.7) | 3,692 | (22.9) | 409 | (23.6) | ||
| ≥30.0 kg/m2 | 124 | (2.7) | 311 | (1.9) | 35 | (2.0) | ||
| Missing data | 74 | (1.6) | 189 | (1.2) | 11 | (0.6) | ||
| METs/day score | ||||||||
| mean (SD) | 34.4 | (6.6) | 34.8 | (6.7) | 35.4 | (6.7) | <0.001 | |
| n (%) | First tertile (range: 23.05–30.10) | 1,211 | (26.8) | 4,412 | (27.4) | 441 | (25.5) | |
| Second tertile (range: 31.85–37.45) | 1,751 | (38.8) | 5,861 | (36.4) | 592 | (34.2) | ||
| Third tertile (range: 37.90–46.25) | 1,360 | (30.1) | 5,334 | (33.1) | 651 | (37.6) | ||
| Missing data | 196 | (4.3) | 519 | (3.2) | 49 | (2.8) | ||
| Smoking status, n (%) | ||||||||
| Never | 2,950 | (65.3) | 9,136 | (56.7) | 890 | (51.4) | <0.001 | |
| Former | 396 | (8.8) | 1,982 | (12.3) | 219 | (12.6) | ||
| 1–19 cigarettes/day | 365 | (8.1) | 1,389 | (8.6) | 154 | (8.9) | ||
| ≥20 cigarettes/day | 771 | (17.1) | 3,478 | (21.6) | 450 | (26.0) | ||
| Missing data | 36 | (0.8) | 141 | (0.9) | 20 | (1.2) | ||
| Alcohol intake, n (%) | ||||||||
| Non-drinker | 2,577 | (57.0) | 7,656 | (47.5) | 752 | (43.4) | <0.001 | |
| Former | 94 | (2.1) | 307 | (1.9) | 28 | (1.6) | ||
| Occasional | 229 | (5.1) | 1,031 | (6.4) | 104 | (6.0) | ||
| 1–<150 g/week | 618 | (13.7) | 2,924 | (18.1) | 338 | (19.5) | ||
| 150–<300 g/week, | 444 | (9.8) | 1,864 | (11.6) | 232 | (13.4) | ||
| ≥300 g/week | 344 | (7.6) | 1,423 | (8.8) | 187 | (10.8) | ||
| Missing data | 212 | (4.7) | 921 | (5.7) | 92 | (5.3) | ||
| Medical history of hypertension, n (%) | ||||||||
| No | 3,616 | (80.0) | 13,217 | (82.0) | 1,439 | (83.0) | 0.01 | |
| Yes | 856 | (19.0) | 2,785 | (17.3) | 278 | (16.0) | ||
| Missing data | 46 | (1.0) | 124 | (0.8) | 16 | (0.9) | ||
| Medication use for hypertension, n (%) | ||||||||
| No | 3,752 | (83.1) | 13,753 | (85.3) | 1,490 | (86.0) | <0.001 | |
| Yes | 754 | (16.7) | 2,308 | (14.3) | 236 | (13.6) | ||
| Missing data | 12 | (0.3) | 65 | (0.4) | 7 | (0.4) | ||
| Medical history of diabetes, n (%) | ||||||||
| No | 4,256 | (94.2) | 15,211 | (94.3) | 1,626 | (93.8) | 0.49 | |
| Yes | 216 | (4.8) | 791 | (4.9) | 91 | (5.3) | ||
| Missing data | 46 | (1.0) | 124 | (0.8) | 16 | (0.9) | ||
| Medication use for diabetes, n (%) | ||||||||
| No | 4,413 | (97.7) | 15,782 | (97.9) | 1,699 | (98.0) | 0.41 | |
| Yes | 91 | (2.0) | 275 | (1.7) | 26 | (1.5) | ||
| Missing data | 14 | (0.3) | 69 | (0.4) | 8 | (0.5) | ||
| Medical history of hypercholesterolemia, n (%) | ||||||||
| No | 4,375 | (96.8) | 15,588 | (96.7) | 1,658 | (95.7) | 0.03 | |
| Yes | 97 | (2.2) | 414 | (2.6) | 59 | (3.4) | ||
| Missing data | 46 | (1.0) | 124 | (0.8) | 16 | (0.9) | ||
| Medication use for hypercholesterolemia, n (%) | ||||||||
| No | 4,418 | (97.8) | 15,704 | (97.4) | 1,675 | (96.7) | 0.14 | |
| Yes | 86 | (1.9) | 354 | (2.2) | 50 | (2.9) | ||
| Missing data | 14 | (0.3) | 68 | (0.4) | 8 | (0.5) | ||
| Job status, n (%) | ||||||||
| Unemployed | 999 | (22.1) | 3,296 | (20.4) | 329 | (19.0) | 0.009 | |
| Employed | 3,471 | (76.8) | 12,692 | (78.7) | 1,394 | (80.4) | ||
| Missing data | 48 | (1.1) | 138 | (0.9) | 10 | (0.6) | ||
| Living alone, n (%) | ||||||||
| No | 4,385 | (97.1) | 15,625 | (96.9) | 1,656 | (95.6) | <0.001 | |
| Yes | 116 | (2.6) | 483 | (3.0) | 75 | (4.3) | ||
| Missing data | 17 | (0.4) | 18 | (0.1) | 2 | (0.1) | ||
| Perceived mental stress, n (%) | ||||||||
| Low | 692 | (15.3) | 2,598 | (16.1) | 357 | (20.6) | <0.001 | |
| Moderate | 2,800 | (62.0) | 10,415 | (64.6) | 923 | (53.3) | ||
| High | 1,007 | (22.3) | 3,044 | (18.9) | 445 | (25.7) | ||
| Missing data | 19 | (0.4) | 69 | (0.4) | 8 | (0.5) | ||
| Type A characteristics, n (%) | ||||||||
| Low | 1,085 | (24.0) | 3,254 | (20.2) | 349 | (20.1) | <0.001 | |
| Moderate | 1,743 | (38.6) | 6,543 | (40.6) | 519 | (30.0) | ||
| High | 647 | (14.3) | 2,624 | (16.3) | 290 | (16.7) | ||
| Very high | 752 | (16.6) | 3,054 | (18.9) | 504 | (29.1) | ||
| Missing data | 291 | (6.4) | 651 | (4.0) | 71 | (4.1) | ||
| Number of friends, n (%) | ||||||||
| No friends | 1,245 | (27.6) | 1,913 | (11.9) | 122 | (7.0) | <0.001 | |
| 1–3 friends | 2,446 | (54.1) | 9,035 | (56.0) | 648 | (37.4) | ||
| ≥4 friends | 744 | (16.5) | 4,992 | (31.0) | 947 | (54.7) | ||
| Missing data | 83 | (1.8) | 186 | (1.2) | 16 | (0.9) | ||
METs, metabolic equivalents; SD, standard deviation.
During the median 11.0 years of follow-up starting 12 years after exposure assessment, we confirmed 3,095 cases of disabling dementia. Having hobbies was inversely associated with the risk of incident disabling dementia (Table 2). Compared with “having no hobbies”, multivariable-adjusted HRs of incident disabling dementia were 0.81 (95% CI, 0.75–0.88) for “having a hobby” and 0.77 (95% CI, 0.66–0.89) for “having many hobbies” after adjustment for age, sex, lifestyle factors, and medical history. A further adjustment for psychosocial factors did not substantially changes the association. In the stratified analysis by age, the inverse association was not significantly different between the participants aged 40–64 years and those 65–69 years at the questionnaire survey (P for interaction: 0.32 for having a hobby and 0.42 for having many hobbies).
Table 2. Hazard ratios (HRs) and 95% confidence intervals (CIs) for incidence disabling dementia according to hobby engagement categories among Japanese aged 40–69 years.
| Hobby categories | ||||
|
| ||||
| Having no hobbies |
Having a hobby |
Having many hobbies |
||
| Total | ||||
| Person-years | 41,483 | 153,232 | 16,340 | |
| Number at risk | 4,518 | 16,126 | 1,733 | |
| Number of cases | 784 | 2,099 | 212 | |
| aModel 1 HR (95% CI) | reference | 0.79 (0.73–0.86) |
0.75 (0.65–0.88) |
|
| bModel 2 HR (95% CI) | reference | 0.81 (0.75–0.88) |
0.77 (0.66–0.89) |
|
| cModel 3 HR (95% CI) | reference | 0.82 (0.75–0.89) |
0.78 (0.67–0.91) |
|
| 40–64 years in 1993–1994 | ||||
| Person-years | 37,636 | 141,642 | 15,113 | |
| Number at risk | 3,898 | 14,344 | 1,553 | |
| Number of cases | 451 | 1,294 | 140 | |
| aModel 1 HR (95% CI) | reference | 0.77 (0.69–0.85) |
0.78 (0.65–0.94) |
|
| bModel 2 HR (95% CI) | reference | 0.79 (0.70–0.88) |
0.82 (0.68–0.99) |
|
| cModel 3 HR (95% CI) | reference | 0.79 (0.71–0.88) |
0.83 (0.68–1.01) |
|
| └0.79 (0.71–0.89)┘ | ||||
| 65–69 years in 1993–1994 | ||||
| Person-years | 3,847 | 11,590 | 1,226 | |
| Number at risk | 620 | 1,782 | 180 | |
| Number of cases | 333 | 805 | 72 | |
| aModel 1 HR (95% CI) | reference | 0.83 (0.73–0.94) |
0.70 (0.54–0.90) |
|
| bModel 2 HR (95% CI) | reference | 0.83 (0.73–0.95) |
0.68 (0.53–0.89) |
|
| cModel 3 HR (95% CI) | reference | 0.84 (0.73–0.96) |
0.68 (0.52–0.90) |
|
| └0.82 (0.72–0.94)┘ | ||||
aAdjusted for age in 1993–1994 and sex.
bAdjusted further for body mass index, smoking status, alcohol intake, total physical activity, history of hypertension, diabetes, and hypercholesterolemia.
cAdjusted further for living alone, job status, perceived mental stress, type A characteristics, and number of friends.
Regarding disabling dementia subtypes, an inverse association of having hobbies was observed for disabling dementia without a history of stroke, but not for post-stroke disabling dementia (Table 3). Model 3 HRs of incident disabling dementia without a history of stroke were 0.77 (95% CI, 0.68–0.88) for “having a hobby” and 0.77 (95% CI, 0.60–0.98) for “having many hobbies”. After stratification by age at the questionnaire survey, the inverse association for disabling dementia without a history of stroke was not significantly different between age categories. For having hobbies combining two hobby engagement categories compared to “having no hobbies” group, multivariable-adjusted HRs of disabling dementia without a history of stroke were 0.71 (95% CI, 0.60–0.85) for participants aged 40–64 years and 0.82 (95% CI, 0.68–0.99) for those aged 65–69 years (P for interaction = 0.09).
Table 3. Hazard ratios (HRs) and 95% confidence intervals (CIs) for incidence disabling dementia subtypes according to hobby engagement categories among Japanese aged 40–69 years.
| Hobby categories | |||||
|
| |||||
| Having no hobbies |
Having a hobby |
Having many hobbies |
|||
| Disabling dementia without a history of stroke | |||||
| Total | |||||
| Person-years | 28,317 | 103,176 | 11,017 | ||
| Number at risk | 4,518 | 16,126 | 1,733 | ||
| Number of cases | 365 | 878 | 90 | ||
| aModel 1 HR (95% CI) | reference | 0.75 (0.66–0.84) |
0.74 (0.59–0.93) |
||
| bModel 2 HR (95% CI) | reference | 0.76 (0.67–0.86) |
0.76 (0.60–0.95) |
||
| cModel 3 HR (95% CI) | reference | 0.77 (0.68–0.88) |
0.77 (0.60–0.98) |
||
| └0.77 (0.68–0.88)┘ | |||||
| 40–64 years in 1993–1994 | |||||
| Person-years | 25,270 | 94,140 | 10,101 | ||
| Number at risk | 3,898 | 14,344 | 1,553 | ||
| Number of cases | 196 | 476 | 53 | ||
| aModel 1 HR (95% CI) | reference | 0.67 (0.57–0.80) |
0.70 (0.51–0.95) |
||
| bModel 2 HR (95% CI) | reference | 0.69 (0.59–0.82) |
0.75 (0.55–1.01) |
||
| cModel 3 HR (95% CI) | reference | 0.71 (0.60–0.84) |
0.75 (0.55–1.04) |
||
| └0.71 (0.60–0.85)┘ | |||||
| 65–69 years in 1993–1994 | |||||
| Person-years | 3,047 | 9,036 | 916 | ||
| Number at risk | 620 | 1,782 | 180 | ||
| Number of cases | 169 | 402 | 37 | ||
| aModel 1 HR (95% CI) | reference | 0.84 (0.70–1.00) |
0.78 (0.55–1.12) |
||
| bModel 2 HR (95% CI) | reference | 0.84 (0.70–1.01) |
0.75 (0.52–1.08) |
||
| cModel 3 HR (95% CI) | reference | 0.83 (0.69–0.998) |
0.75 (0.52–1.10) |
||
| └0.82 (0.68–0.99)┘ | |||||
| Post-stroke disabling dementia | |||||
| Total | |||||
| Number of cases | 114 | 318 | 34 | ||
| aModel 1 HR (95% CI) | reference | 0.83 (0.67–1.02) |
0.84 (0.57–1.23) |
||
| bModel 2 HR (95% CI) | reference | 0.87 (0.70–1.08) |
0.87 (0.59–1.28) |
||
| cModel 3 HR (95% CI) | reference | 0.89 (0.71–1.11) |
0.93 (0.62–1.38) |
||
| └0.89 (0.71–1.11)┘ | |||||
| 40–64 years in 1993–1994 | |||||
| Number of cases | 60 | 191 | 18 | ||
| aModel 1 HR (95% CI) | reference | 0.83 (0.62–1.12) |
0.70 (0.42–1.19) |
||
| bModel 2 HR (95% CI) | reference | 0.88 (0.65–1.18) |
0.74 (0.44–1.26) |
||
| cModel 3 HR (95% CI) | reference | 0.89 (0.66–1.20) |
0.76 (0.44–1.32) |
||
| └0.88 (0.65–1.18)┘ | |||||
| 65–69 years in 1993–1994 | |||||
| Number of cases | 54 | 127 | 16 | ||
| aModel 1 HR (95% CI) | reference | 0.81 (0.59–1.12) |
1.06 (0.60–1.85) |
||
| bModel 2 HR (95% CI) | reference | 0.84 (0.60–1.16) |
1.01 (0.57–1.78) |
||
| cModel 3 HR (95% CI) | reference | 0.85 (0.61–1.19) |
1.09 (0.60–1.98) |
||
| └0.87 (0.62–1.21)┘ | |||||
aAdjusted for age in 1993–1994 and sex.
bAdjusted further for body mass index, smoking status, alcohol intake, total physical activity, history of hypertension, diabetes, and hypercholesterolemia.
cAdjusted further for living alone, job status, perceived mental stress, type A characteristics, and number of friends.
After excluding participants with missing data of confounding variables, the results did not change materially (eTable 1 and eTable 2).
DISCUSSION
In the current study, hobby engagement in the age range of 40–69 years (mean, 53 years) was associated with a lower risk of disabling dementia among Japanese men and women during 12 to 23 years after hobby assessment. Further, the inverse association was apparent for disabling dementia without a history of stroke, but not for post-stroke disabling dementia. The association between hobby engagement and disabling dementia were similar between middle (40–64 years) and older ages (65–69 years) at the questionnaire survey.
In the Monongahela Valley Independent Elders Survey project of 942 individuals in the United States aged ≥65 years with an average of 6 years of follow-up, a longer time commitment to hobbies was associated with a lower risk of dementia.8 A 6-year follow-up for Japanese men and women aged ≥65 years in the Japan Gerontological Evaluation Study (JAGES) showed that several types of hobbies, such as ground golf and travel, were associated with a lower risk of dementia.6 Moreover, the UK Million Women Study showed that the participation in groups for art, craft, or music at a mean age of 60 years and reading at a mean age of 64 years were associated with a lower risk of dementia in 0–4 and 5–9 years follow-up, but not in ≥10 years follow-up.22 The Betula prospective cohort study of individuals aged ≥65 years also reported that leisure activity was associated with a lower risk of dementia in the first period (1–5 years after baseline), but not in the second and third periods (6–10 and 11–15 year after baseline).5 The findings from the above two studies were different from ours probably because of different timing for exposure determination in late life and mid-life. Our finding in mid-life corroborated the result from the Gothenburg H70 Birth Cohort Study of 800 Swedish women aged 38–54 years, which reported that cognitive and physical activity in mid-life were associated with a lower risk of dementia in a mean of 44 years of follow-up.13
In the present study, we found that hobby engagement was inversely associated with the risk of disabling dementia without a history of stroke, which may correspond to non-vascular type dementia, most likely Alzheimer’s disease. In most previous studies, cognitive activity was associated with a lower risk of Alzheimer’s disease,2–4,7,13 while only one study reported a protective effect of cognitive activity on vascular dementia and mixed dementia.3 On the other hand, physical activity was inversely associated with a risk of Alzheimer’s disease,2,10 vascular dementia,12 dementia with cerebrovascular disease,13 and mixed dementia.13
Several possible mechanisms can be addressed for the protective effect of hobby engagement on dementia. First, hobby engagement, as an enjoyable leisure activity at age of 19–89 years, was correlated with enhanced life engagement (purpose in life),1 which was associated with a lower risk of dementia14 partly due to lower levels of inflammatory markers.23 Furthermore, high average cognitive activity across ages of 6, 12, 18, and 40 years was associated with lower β-amyloid accumulation measured using positron emission tomography among people aged 50 years or older,24 suggesting that cognitive activity from the early life makes cognitive reserve and prevents dementia. On the other hand, high physical activity (such as ≥2.5 hours of moderate to vigorous physical activity/week and participation in leisure time physical activity at ≥2 times/week) in mid-life was associated with lower levels of inflammatory markers25 and a larger volume of gray matter26 in late life compared to the lower physical activity. Further, physical activity in old age of 60–95 years was associated with lower β-amyloid accumulation.27 Physical activity decreases the risk of the development of hypertension, diabetes, and obesity,28–30 all of which were risk factors for the development of dementia.31
The strengths of this study are the large sample size, minimizing the probability of chance findings. Incident disabling dementia was identified using routinely collected government data with validated methods, which reduced the possibility of reporting error of outcome.
This study has several limitations. First, we had no information about the cognitive function and a history of dementia at the time of the questionnaire survey in 1993–1994. Cognitive decline and dementia may lead to reduced hobby engagement; therefore, reverse causation cannot be inevitable. In the present study, however, dementia outcome was assessed 12 years after the exposure assessment, and the impact of reverse causation may be small. Second, we could not discuss the specific associations of the actual number, types, frequency, and intensity of hobbies with the risk of disabling dementia because we did not ask for these information. Previous studies showed that the number of hobbies,6 time commitment to hobbies,8 and the intensity of cognitive and physical activity13 were associated with a lower risk of dementia. Third, we assessed the hobby only at the questionnaire survey in 1993–1994, so we could not take into account the changes in hobby engagement during the follow-up. Some of the participants may quit or start hobbies during the follow-up. For example, starting a new hobby in late life could lead to lowering the dementia risk, attenuating the association between hobby engagement and the dementia risk. Fourth, we did not have a precise diagnosis of dementia subtypes, including Alzheimer’s dementia, vascular dementia, and mixed dementia, although we found an inverse association between hobby engagement and the risk of disabling dementia without a history of stroke, a surrogate outcome for Alzheimer’s disease. Finally, residual and unmeasured confounding factors may exist, such as education levels, the presence of psychiatric disorders, and use of psychiatric medications.
In conclusion, hobby engagement in both mid-life and late life was associated with a lower risk of disabling dementia without a history of stroke among the Japanese population.
ACKNOWLEDGMENTS
This study was supported by a Grant-in-Aid from the National Cancer Center Research and Development Fund (since 2011), Grant-in-Aid for Cancer Research from the Ministry of Health, Labour and Welfare of Japan (from 1989 to 2010), and the Japan Society for the Promotion of Science (Scientific Research (B) grant numbers 16H05246 and 21H03194), and the Osaka University International Joint Research Promotion Program at University College London.
The authors thank all the staff members for their valuable help in conducting the baseline survey and follow-up. The JPHC members are listed at the following site: https://epi.ncc.go.jp/en/jphc/781/index.html.
Data Availability: For information on how to submit an application for gaining access to JPHC data and/or biospecimens, please follow the instructions at https://epi.ncc.go.jp/en/jphc/805/8155.html.
Conflicts of interest: None of declared.
SUPPLEMENTARY MATERIAL
The following is the supplementary data related to this article:
eTable 1. Hazard ratios (HRs) and 95% confidence intervals (CIs) for incidence disabling dementia according to hobby engagement categories among Japanese aged 40–69 years excluding individuals with missing data of confounding variables
eTable 2. Hazard ratios (HRs) and 95% confidence intervals (CIs) for incidence disabling dementia subtypes according to hobby engagement categories among Japanese aged 40–69 years excluding individuals with missing data of confounding variables
REFERENCES
- 1.Pressman SD, Matthews KA, Cohen S, et al. Association of enjoyable leisure activities with psychological and physical well-being. Psychosom Med. Sep 2009;71(7):725–732. 10.1097/PSY.0b013e3181ad7978 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Scarmeas N, Levy G, Tang MX, Manly J, Stern Y. Influence of leisure activity on the incidence of Alzheimer’s disease. Neurology. Dec 2001;57(12):2236–2242. 10.1212/WNL.57.12.2236 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Verghese J, Lipton RB, Katz MJ, et al. Leisure activities and the risk of dementia in the elderly. N Engl J Med. Jun 2003;348(25):2508–2516. 10.1056/NEJMoa022252 [DOI] [PubMed] [Google Scholar]
- 4.Akbaraly TN, Portet F, Fustinoni S, et al. Leisure activities and the risk of dementia in the elderly: results from the Three-City Study. Neurology. Sep 2009;73(11):854–861. 10.1212/WNL.0b013e3181b7849b [DOI] [PubMed] [Google Scholar]
- 5.Sörman DE, Sundström A, Rönnlund M, Adolfsson R, Nilsson LG. Leisure activity in old age and risk of dementia: a 15-year prospective study. J Gerontol B Psychol Sci Soc Sci. Jul 2014;69(4):493–501. 10.1093/geronb/gbt056 [DOI] [PubMed] [Google Scholar]
- 6.Ling L, Tsuji T, Nagamine Y, Miyaguni Y, Kondo K. [Types and number of hobbies and incidence of dementia among older adults: a six-year longitudinal study from the Japan Gerontological Evaluation Study (JAGES)]. Nihon Koshu Eisei Zasshi. 2020;67(11):800–810. 10.11236/jph.67.11_800 [DOI] [PubMed] [Google Scholar]
- 7.Wilson RS, Mendes De Leon CF, Barnes LL, et al. Participation in cognitively stimulating activities and risk of incident Alzheimer disease. JAMA. Feb 2002;287(6):742–748. 10.1001/jama.287.6.742 [DOI] [PubMed] [Google Scholar]
- 8.Hughes TF, Chang CC, Vander Bilt J, Ganguli M. Engagement in reading and hobbies and risk of incident dementia: the MoVIES project. Am J Alzheimers Dis Other Demen. Aug 2010;25(5):432–438. 10.1177/1533317510368399 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Almeida OP, Yeap BB, Alfonso H, Hankey GJ, Flicker L, Norman PE. Older men who use computers have lower risk of dementia. PLoS One. 2012;7(8):e44239. 10.1371/journal.pone.0044239 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kishimoto H, Ohara T, Hata J, et al. The long-term association between physical activity and risk of dementia in the community: the Hisayama Study. Eur J Epidemiol. Mar 2016;31(3):267–274. 10.1007/s10654-016-0125-y [DOI] [PubMed] [Google Scholar]
- 11.Lee ATC, Richards M, Chan WC, Chiu HFK, Lee RSY, Lam LCW. Association of daily intellectual activities with lower risk of incident dementia among older Chinese adults. JAMA Psychiatry. Jul 2018;75(7):697–703. 10.1001/jamapsychiatry.2018.0657 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hansson O, Svensson M, Gustavsson AM, et al. Midlife physical activity is associated with lower incidence of vascular dementia but not Alzheimer’s disease. Alzheimers Res Ther. Oct 2019;11(1):87. 10.1186/s13195-019-0538-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Najar J, Östling S, Gudmundsson P, et al. Cognitive and physical activity and dementia: a 44-year longitudinal population study of women. Neurology. Mar 2019;92(12):e1322–e1330. 10.1212/WNL.0000000000007021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Boyle PA, Buchman AS, Barnes LL, Bennett DA. Effect of a purpose in life on risk of incident Alzheimer disease and mild cognitive impairment in community-dwelling older persons. Arch Gen Psychiatry. Mar 2010;67(3):304–310. 10.1001/archgenpsychiatry.2009.208 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tsugane S, Sawada N. The JPHC study: design and some findings on the typical Japanese diet. Jpn J Clin Oncol. Sep 2014;44(9):777–782. 10.1093/jjco/hyu096 [DOI] [PubMed] [Google Scholar]
- 16.Long-term Care Insurance in Japan [Internet]. Ministry of Health, Labor and Welfare of Japan [cited 2021 Apr 18]. Available from: https://www.mhlw.go.jp/english/topics/elderly/care/index.html.
- 17.Tamiya N, Noguchi H, Nishi A, et al. Population ageing and wellbeing: lessons from Japan’s long-term care insurance policy. Lancet. Sep 2011;378(9797):1183–1192. 10.1016/S0140-6736(11)61176-8 [DOI] [PubMed] [Google Scholar]
- 18.Inoue M, Yamamoto S, Kurahashi N, et al. ; Japan Public Health Center-based Prospective Study Group . Daily total physical activity level and total cancer risk in men and women: results from a large-scale population-based cohort study in Japan. Am J Epidemiol. Aug 2008;168(4):391–403. 10.1093/aje/kwn146 [DOI] [PubMed] [Google Scholar]
- 19.Ikeda A, Iso H, Kawachi I, Inoue M, Tsugane S; JPHC Study Group . Type A behaviour and risk of coronary heart disease: the JPHC Study. Int J Epidemiol. Dec 2008;37(6):1395–1405. 10.1093/ije/dyn124 [DOI] [PubMed] [Google Scholar]
- 20.Noda H, Yamagishi K, Ikeda A, Asada T, Iso H. Identification of dementia using standard clinical assessments by primary care physicians in Japan. Geriatr Gerontol Int. May 2018;18(5):738–744. 10.1111/ggi.13243 [DOI] [PubMed] [Google Scholar]
- 21.Yamagishi K, Iso H, Kokubo Y, et al. ; JPHC Study Group . Dietary intake of saturated fatty acids and incident stroke and coronary heart disease in Japanese communities: the JPHC Study. Eur Heart J. Apr 2013;34(16):1225–1232. 10.1093/eurheartj/eht043 [DOI] [PubMed] [Google Scholar]
- 22.Floud S, Balkwill A, Sweetland S, et al. Cognitive and social activities and long-term dementia risk: the prospective UK Million Women Study. Lancet Public Health. Feb 2021;6(2):e116–e123. 10.1016/S2468-2667(20)30284-X [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Engelhart MJ, Geerlings MI, Meijer J, et al. Inflammatory proteins in plasma and the risk of dementia: the rotterdam study. Arch Neurol. May 2004;61(5):668–672. 10.1001/archneur.61.5.668 [DOI] [PubMed] [Google Scholar]
- 24.Landau SM, Marks SM, Mormino EC, et al. Association of lifetime cognitive engagement and low β-amyloid deposition. Arch Neurol. May 2012;69(5):623–629. 10.1001/archneurol.2011.2748 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hamer M, Sabia S, Batty GD, et al. Physical activity and inflammatory markers over 10 years: follow-up in men and women from the Whitehall II cohort study. Circulation. Aug 2012;126(8):928–933. 10.1161/CIRCULATIONAHA.112.103879 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Rovio S, Spulber G, Nieminen LJ, et al. The effect of midlife physical activity on structural brain changes in the elderly. Neurobiol Aging. Nov 2010;31(11):1927–1936. 10.1016/j.neurobiolaging.2008.10.007 [DOI] [PubMed] [Google Scholar]
- 27.Brown BM, Peiffer JJ, Taddei K, et al. Physical activity and amyloid-β plasma and brain levels: results from the Australian Imaging, Biomarkers and Lifestyle Study of Ageing. Mol Psychiatry. Aug 2013;18(8):875–881. 10.1038/mp.2012.107 [DOI] [PubMed] [Google Scholar]
- 28.Huai P, Xun H, Reilly KH, Wang Y, Ma W, Xi B. Physical activity and risk of hypertension: a meta-analysis of prospective cohort studies. Hypertension. Dec 2013;62(6):1021–1026. 10.1161/HYPERTENSIONAHA.113.01965 [DOI] [PubMed] [Google Scholar]
- 29.Aune D, Norat T, Leitzmann M, Tonstad S, Vatten LJ. Physical activity and the risk of type 2 diabetes: a systematic review and dose-response meta-analysis. Eur J Epidemiol. Jul 2015;30(7):529–542. 10.1007/s10654-015-0056-z [DOI] [PubMed] [Google Scholar]
- 30.Bell JA, Hamer M, Batty GD, Singh-Manoux A, Sabia S, Kivimaki M. Combined effect of physical activity and leisure time sitting on long-term risk of incident obesity and metabolic risk factor clustering. Diabetologia. Oct 2014;57(10):2048–2056. 10.1007/s00125-014-3323-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. Aug 2020;396(10248):413–446. 10.1016/S0140-6736(20)30367-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.