Characteristics of research with older people (over 65 years) in occupational therapy journals, 2013–2017

Abstract

Background: A growing population of older people will require different types of occupational therapy services in the future. For occupational therapists to provide effective services and to optimize care, their practice must rely on high-quality evidence. Research is one important pillar of evidence-based practice. Therefore, it is important to examine the research published in occupational therapy journals, which guides practitioners in their work with clients.

Aim: The overall aim of this study was to review research characteristics in articles with older persons as participants, aged over 65 years, with or without illness, diseases or disabilities, reported in occupational therapy journals during the period 2013–2017.

Materials and methods: Data was collected from peer-reviewed occupational therapy journals and categorized in relation to research characteristics using descriptive statistics.

Results: The findings show that most articles presented basic research, using quantitative design where the sources of data were instruments.

Conclusion: The findings suggest that both qualitative and quantitative articles use appropriate sample sizes. However, descriptions of the studied populations are frequently unclear, which may affect the transferability and generalization of the results.

Significance: In order to support practice, efforts are needed to develop research aims and questions that develop knowledge to embrace more than just basic research.

Keywords:

• Aged
• aging
• research design

Introduction

Today, a demographic shift is occurring worldwide as the number of people over 60 years of age is growing faster than any other age group [1]. This growing population of older people will require different occupational therapy services in the future to support healthy aging [2] and to remain active citizens. Occupational therapy can be cost-effective and can motivate older people to maintain their own health [3].

For occupational therapists to provide effective services and to optimize care, their practice must rely on credible evidence. Evidence-based practice relies on three pillars: the patient’s wishes or experiences, professional expertise, and the best available knowledge [4]. Research is a prerequisite for best available knowledge. Thus, in order to support and develop the practice of occupational therapy, research is needed [5]. How occupational therapists maintain currency with the latest published research has been described previously [6]. Journals specific to occupational therapy are presumed to represent a primary source for practising occupational therapists’ access to research, as these journals are often distributed amongst the members of different occupational therapy associations [7,8]. Criticism has been raised that the research about older people published in occupational therapy journals is of a basic character (i.e. descriptive research that provides knowledge about experiences or a phenomenon but does not necessarily translate the results into guidance for practice) and that there is a need for more research on interventions and their effectiveness [9]. A ten-year-old review of six occupational therapy journals’ characteristics concluded that only 15 percent of all published articles included older people as participants. The majority of these studies took a quantitative approach and the topics reported revolved around instrument testing and development [8].

Literature reviews are used to synthesize existing research, what has been done, what is already known or what gaps exist in knowledge about a certain topic [10]. Literature reviews can also be used to present the characteristics of articles published in a certain research field [11] or to review certain areas or topics of articles published in occupational therapy journals [6,8,9,12–14]. Previous research concerning articles published in the field of occupational therapy has analyzed the references used in publications [6,15,16], or quantified the influence of scholarly activities [17–19]. In recent years, there has been an increased interest in developing research about older people in the field of occupational therapy [9], and both the quality (i.e. more effectiveness studies) and the quantity of research has improved [13]. Despite this increased interest in research about older people and its significance for evidence-based practice, there is no up-to-date review of such research published in international occupational therapy journals focusing on its characteristics in terms of publication characteristics, descriptions of participants, research design, and subject field.

Aim and research questions

The overall aim of this study was to review the research characteristics in articles with older persons as participants, aged over 65 years with and without illness, disease or disabilities, reported in occupational therapy journals during the period 2013–2017. Our specific research questions were:

1. What are the publication characteristics regarding older people in occupational therapy journals (i.e. number of publications and origin of authors)?

2. Which groups of older people have been studied in occupational therapy journals (i.e. descriptions of groups and ages of participants)?

3. What types of research design have been used in research with older people published in occupational therapy journals (i.e. form of research, study design, collection of data)?

4. Which subject fields have been studied in research with older people published in occupational therapy journals?

Methods

Selection of journals

The journals selected for this review are professional journals (i.e. journals that publish discipline-related research [18]) concerning occupational therapy. There are several profession-specific journals relating to occupational therapy but not all have a journal impact factor [17,18]. To include articles based only on journal impact factor, there is a risk that relevant articles concerning occupational therapy will be excluded [18] since scholars in occupational therapy do not always choose journals to publish in based on impact factor [20]. Thus, in this review, journals were selected on the basis of the following criteria: peer-reviewed occupational therapy journals, published in English and accessible through the electronic databases Medline, OTDBase, and Cumulative Index to Nursing and Allied Health Literature (CHINAL) [17]. The selected journals are presented in Table 1.

Table 1. Presentation of selected journals for this review and journal abbreviations.

Selected journals*	Journal abbreviations
American Journal of Occupational Therapy	AJOT
Asian Journal of Occupational Therapy	AsJOT
Australian Occupational Therapy Journal	AOTJ
British Journal of Occupational Therapy	BJOT
Canadian Journal of Occupational Therapy	CJOT
Hong Kong Journal of Occupational Therapy	HKOT
Indian Journal of Occupational Therapy	IJOT
New Zealand Journal of Occupational Therapy	NZOT
Occupational Therapy in Health Care	OTHC
Occupational Therapy in Mental Health	OTMH
Occupational Therapy International	OTI
OTJR; Occupation, Participation and Health	OTJR
Open Journal of Occupational Therapy	OJOT
Physical and Occupational Therapy in Geriatrics	POTG
Scandinavian Journal of Occupational Therapy	SJOT

*The South African Journal of Occupational Therapy, Japanese Journal of Occupational therapy Research, Israel Journal of Occupational therapy, Korean Journal of Occupational Therapy and World Federation of Occupational Therapist Bulletin was searched although no articles met the inclusion criteria.

All the included journals’ guidelines for authors stress that the research should have been conducted in an ethical and responsible manner, and in compliance with relevant codes of experimentation and the requirements of legislation.

Inclusion of articles

Broad inclusion criteria were employed. Empirical research (qualitative, quantitative, and mixed-methods research) published between 2013 and 2017 in which the majority of participants were older than 65 years (when mean age was reported in the articles it had to be 65 years or higher, or participants described as older adults) was included. A five-year period has been used in similar reviews of articles published in occupational therapy journals [8,11]. This five-year period was thought to be suitable to avoid bias in the results due to special issues that could have skewed the results (for example, proportions of published articles, or certain topics). Articles reporting occupational therapists’ experiences or descriptions in relation to working with older people, literature studies, publications ahead of print, systematic reviews, meta analyses, meta syntheses, editorials, and letters to the editor were excluded. In POTG, articles reporting research in the field of physiotherapy were excluded. The exclusion process for articles in POTG based on this criterion was conducted by both authors jointly until consensus was reached.

A manual search was conducted to include relevant articles from the selected journals. All published articles in the selected journals were assessed in relation to the inclusion and exclusion criteria. This search strategy has been used in similar reviews to include articles published in occupational therapy journals [8,9,11–13]. In total, 2960 articles were published in the included journals between 2013 and 2017. Articles with keywords in the titles or abstracts describing the studied population as children, teenagers, or adults of working age (such as children, school pupils, students, teenagers, work, and youth) were excluded from this review.

Articles including keywords related to older people (such as aging, geriatric, elder, elderly, seniors, retirement) but also in relation to diseases/injuries/disabilities (such as dementia disease, hip fractures, low vision, Parkinson’s disease, stroke), interventions (such as fall prevention, hip replacement), or locations (such as assisted living, care homes, nursing facilities) usually correlating to advanced age in either the title or abstract were checked in relation to the age of the participants. If an article could not be excluded based on title or abstract alone, the full-text article was used to identify the age of the participants to determine whether or not it should be included in this review. In total, 257 articles were included. All articles that had been excluded were examined once more to ensure that no relevant articles had been missed. This manual search was conducted by both authors separately and deviances were discussed until a consensus was reached. The complete list of studies included in this review may be obtained from the authors.

Data analysis

Data was extracted from the included articles and compiled in an Excel database file created for this review. This database file contained the data needed in order to review the articles’ research characteristics. This data was then classified according to the subject fields (i.e. the topic defined in the article, such as stroke, driving), groups of older people, age of the participants, research design, data used, and the first authors’ institutional affiliations (i.e. in which country the authors worked). This selection of research characteristics was based on an earlier study [8]. The descriptions for the classifications were based on the terminology of each article (for example, some articles used interventions broadly and were classified as ‘intervention’ whilst others specifically used ‘fall prevention’ and were thus classified as ‘fall prevention’; some articles used activities in general whilst others specified the activity, such as driving, and thus the former were classified as ‘activities’ and the latter as ‘driving’).

The International Classification of Functioning, Disability and Health [ICF] provides a standard language and framework for making research results comparable [21]. The titles, aims and research questions of the included articles were used to link [22] them to the ICF’s components: ‘Body functions and body structures’, ‘Activity and participation’, ‘Environmental factors’, and ‘Personal factors’.

The included articles were also classified in relation to Gutman’s [23] five specific research priorities or research categories: basic research (i.e. research that provides information about disabilities and their impact on functional participation), instrument testing (i.e. articles testing the reliability and validity of assessment instruments), correlational and descriptive articles (i.e. articles that demonstrate the link between occupational engagement and health), effectiveness studies (i.e. articles that support practice) and professional topics (i.e. articles that provide answers to important questions concerning the profession’s growth).

Linking to the ICF’s components was conducted by the first author and the classification in accordance with Gutman’s [23] five specific research priorities was made by the second author. The links to the ICF’s components [21] and the classification in accordance with Gutman’s [23] research priorities were each validated by the other author and discussed until consensus was reached.

Descriptive statistics were used to analyze the data, and in this review we present our findings in frequencies and percentages.

Findings

Characteristics of publications

In total, 2960 articles were published in the included occupational therapy journals between 2013 and 2017. Of these articles, 257 (8.7%) were empirical research in which the majority of participants were older than 65 years or the mean age was reported as 65 years or more. In Table 2, descriptive statistics are presented showing how many studies in the included journals were about older people, and the total number of studies published each year.

Table 2. Number and frequency of included articles according to inclusion criteria and articles published in each journal per year according to scopus or the journals homepage (*).

	2013	2014	2015	2016	2017	Total
AJOT	7/102 = 6.9%	10/131 = 7.6%	9/154 = 5.8%	9/125 = 7.2%	7/103 = 6.8%	42/615 = 6.8%
AOTJ	9/73 = 12.3%	3/76 = 3.9%	5/81 = 6.2%	3/64 = 4.7%	5/87 = 5.7%	25/381 = 6.6%
AsJOT*	0	1/3 = 33.3%	0	1/19 = 5.2%	5/12 = 41.6%	7/34 = 20.5%
BJOT	25/381 = 6.6%	8/101 = 7.9%	12/99 = 12.1%	9/108 = 8.3%	12/86 = 14%	45/483 = 9.3%
CJOT	2/42 = 4.8%	2/38 = 5.3%	2/38 = 5.3%	1/38 = 2.6%	1 /21 = 4.8%	8/177 = 4.5%
HKOT	2/13 = 15.4%	1/13 = 7.7%	2/15 = 13.3%	1/12 = 8.3%	1/14 = 7.1%	7/67 = 10.4%
IJOT*	0/12 = 0%	0/12 = 0%	1/13 = 7.7%	0	0	1/37 = 2.7%
NZJOT*	0/9 = 0%	0/8 = 0%	0/8 = 0%	1/10 = 10%	0/8 = 0%	1/43 = 2.3%
OJOT*	1/16 = 6.2%	3/23 = 13.0%	0/30 = 0%	0/43 = 0%	0/49 = 0%	4/161 = 2.5%
OTMH*	1/24 = 4.2%	0/17 = 0%	0/23 = 0%	1/24 = 4.2%	0/24 = 0%	2/112 = 1.8%
OTHC	1/48 = 2%	2/44 = 4.5%	0/34 = 0%	2/32 = 6.2%	4/33 = 12.1%	9/191 = 4.7%
OTI	4/26 = 15.4%	4/21 = 19%	2/22 = 9.1%	2/40 = 5%	1/24 = 4.2%	13/133 = 9.8%
OTJR*	3/22 = 13.6%	1/20 = 5%	4/24 = 16.7%	2/24 = 8.3%	0/23 = 0%	10/113 = 8.8%
POTG	14/27 = 51.8%	16/31 = 51.6%	8/27 = 29.6%	3/14 = 21.4%	5 /14 = 35.7%	46/113 = 40.7%
SJOT	4/63 = 6.3%	12/71 = 16.9%	7/65 = 10.5%	5/49 = 10.2%	9/ 52 = 17.3%	37/300 = 12.3%

As can be seen in Table 2, there is a fluctuation in frequencies and percentages of the proportion of articles including older people as participants published in the journals over the five-year period in question. Over this period, the journal with the highest proportion of articles including older people is POTG (40.7%), followed by AsJOT (20.5%), SJOT (12.3%), HKOT (10.4%), OTI (9.8%), and BJOT (9.3%).

The first authors of the articles represented 26 countries altogether. Most first authors were from the United States of America (78/30.4%), Australia (38/14.8%), Sweden (37/14.4%), Canada (30/11.7%) and Japan (21/8.6%). These five countries accounted for almost 80% of the articles. The United Kingdom (15/5.8%), Norway (7/2.9%), South Korea (4/1.6%) Denmark and Finland (3/1.2%) had more than two first-authored articles whilst the remaining 15 countries had one or two first-authored articles.

Description and age of participants

The age of participants in the included articles was given as a mean, a range or both. In 202 articles, mean age was given and varied between 65 and 91 years of age, with a median of 74.4 years. In 158 articles, a range of ages was given. The youngest participant was 17 years old (range 17–94) and the oldest was 103 years old (range 84–103). In 81 of the included articles, participants under the age of 65 were included, although the mean age was over 65. In four of the included articles, no age was given; the participants were described as older adults or aged.

Most participants were described according to a diagnosis (115/44.7%). The diagnosis used most frequently to categorize participants was stroke (37/14.4%), followed by a dementia diagnosis (22/8.6%), mild cognitive impairments (11/4.2%), or an orthopedic diagnosis (11/4.2%). Many articles gave vague or imprecise definitions of the participants, such as older persons, community-dwelling, older drivers, or elderly persons with or without impairments (110/42.8%). Another way to categorize the participants was to describe them as institutionalized or patients on a specific ward (23/8.9%) in different care homes or nursing home facilities. The remaining articles described the participants as mixed (2), in rehab (6) or users of technical aids (1).

The number of participants in the articles varied from one to 27,131. In the qualitative articles, the number of included participants ranges from 1–149, while in the quantitative articles the range is 1–27,131, and in the mixed-methods articles the numbers range from 1–506. In Table 3, the number of participants is presented in relation to research design.

Table 3. Number of participants in relation to research design.

	No art	Range	Q1	Median	Q3
Basic research	130	1–27,131	10	30	78
Qualitative	47	3–60	6	10	17
Quantitative	66	1–27131	33	63	176
Mixed method	19	1–506	3	10	40
Correlational and descriptive	39	5–506	39	80	131
Qualitative	9	5–149	8	13	73
Quantitative	26	14–506	61	98	143
Mixed methods	4	39–456
Effectiveness	47	1–12,094	11	20	55
Qualitative	2	2–5
Quantitative	38	1–12,094	11	21	118
Mixed Method	7	14–52
Instrument testing	40	5–1184	27	96	207
Qualitative	1	5
Quantitative	37	5–1184	29	96	207
Mixed methods	1	27
Professional topic	1	35
Qualitative	1	35
Total	n = 257	1–27,131	12	37	118

Research categories, study design and collection of data in the included articles

The classification of research categories according to Gutman [23] revealed that 130 (50.5%) articles were basic research that provided information about disabilities and their impact on functional participation. Forty (40/15.5%) were articles testing the reliability and validity of instruments, and 39 (15.1%) were correlational and descriptive articles demonstrating the link between occupational engagement and health. Forty-seven (47/18.2%) were effectiveness articles that supported practice, and finally one article (1/0.4%) provided answers to important questions concerning the profession’s growth. Of the 257 articles, 167 (64.9%) employed a quantitative design, 60 (23.3%) employed a qualitative design, and 30 (11.6%) used a mixed-methods design. In Table 4, the study design is presented in relation to the category of research priorities according to Gutman [23]. Most articles (65/25.2%) were basic research using a quantitative design, followed by basic research using a qualitative design (47/18.2%).

Table 4. Cross table of study design and category of research [23].

	Basic research	Correlational and descriptive	Effectiveness	Instrument testing	Professional topic	Total (n = 257 articles)
Qualitative	47	9	2	1	1	60
Quantitative	66	26	38	37		167
Mixed methods	17	4	7	2		30
Total	130	39	47	40	1	257

The method for data collection in almost half of the included articles was based on instruments (119/46.3%), while interviews in different formats (i.e. semi-structured interviews, focus-group interviews, in-depth interviews) accounted for almost a quarter (63/24.5%). Questionnaires/surveys were used in 27 (10.5%) articles. Other methods of collecting data were [medical] records (14/5.4%), observations (12/4.6%), videos/photos (5/1.9%), tests (10/3.8%), diaries (3/1.2%), and other types of data (4/1.5%). One fifth of the articles (55/21.4%) used triangulation involving more than one source of data. Combining research categories [23], study design and collection of data shows that basic research using quantitative methods most frequently used instruments for data collection, whilst basic research using qualitative methods used interviews of different kinds.

Subject fields of articles

The classification of subjects covered by the articles resulted in interventions being the largest subject field, with 42 articles (16.3%), followed by activities (i.e. ADL, everyday activities) in 36 articles (14%), and instrument development and testing in 35 articles (13.6%). Other subjects with more than ten articles per area were: driving (19/7.3%), falls and fall prevention (17/6.6%), function (15/5.8%), technology use (13/5.1%), and participation (11/4.2%). In all, 47 subject fields were identified: 18 with two or more articles and 29 with one article each. In Table 5, the 10 largest subject fields are presented.

Table 5. Classification of subject fields linked to ICF’s components [21].

	B	D	E	S	B;D	B;E	B;PF	B;S	D;E	D;PF	E;PF	B;D;E	NC	Total
Intervention	2	10	7		4	6			8	2		1	2	41
Activity		17			7		1		7	3		1		34
Instrument testing	9	9	2		3	2			3				7	32
Driving		12			3				4					17
Falls	2	6	2		2				3			1	1	15
Function	5			1	5	1	1	1	1					15
Technology use			3						8			2		13
Participation	1	5			1				1	3				10
Occupational performance		3			3				2					8
Technical aids			4									1		5
Remaining articles	6	13	14		3	2			9	1	2		6	54
Total	25	75	32	1	31	11	2	1	46	9	2	6	16	257

B: body function; D: activity and participation; E: environment; S: body structure; PF: personal factors; NC: not covered.

Classification according to the ICF [21] revealed that most articles (75/29.2%) were linked as activity and participation (D). The ICF environment (E) constituted 32 articles (12.4%), body function (B) 25 articles (9.7%) and body structure (S) one article (0.4%). Several of the articles were linked to two components: activity and participation, and environment (D;E) 46 articles (17.8%), body function, and activity and participation (B;D) 31 articles (12.1%), body function, and environment (B;E) 11 articles (4.2%), activity and participation, and personal factors (D;PF) nine articles (3.5%), body function, and personal factors (B;PF) two articles (0.8%), environment, and personal factors (E;PF) two articles (0.8%), body function, and body structure (B;S) one article (0.4%). Six articles were linked to three components in ICF, body function, activity and participation, and environment (B;D;E, 2.3%). Lastly, 16 articles were classified as not covered (NC) (6.2%).

Combining the subject fields of the article with the linking to the ICF components revealed that all but one articles in the subject field activity were linked to activity and participation. Function was linked in all but one case to body function or structure. Articles linked to “not covered by the ICF” were mostly in the subject field of instrument testing (7 articles). Articles about interventions were linked in all but one case to activity and participation and/or environment.

Combining subject field with category of research according to Gutman [23] revealed that most articles (28) in the subject field of instruments was in the category of instrument testing, representing 70% in that category (instrument testing). One third (16/34%) of the effectiveness studies related to interventions, which represents 38.1% of the intervention studies. Furthermore, the subject of driving features 12 (63.2%) articles as basic research, three (3/15.7%) as effectiveness, three (3/15.7%) as instrument testing, and one (1/5.2%) as correlational and descriptive studies. Subject areas with a high proportion of basic research were: occupational performance (87.5%), function (66.7%), technology use (61.5%), falls (58.8%), activity (58.3%) and interventions (45.2%). See Table 6.

Table 6. Cross table of subject fields and categories of research [23] covered by the articles.

	Basic research	Correlational and descriptive	Effectiveness	Instrument testing	Professional topic	Total (n = 257 articles)
Interventions	19	3	16	4		42
Activity	21	5	7	3		36
Instruments	5	1	1	28		35
Driving	12	1	3	3		19
Falls	10	4	3			17
Function	10	2	3			15
Technology use	8	4	1			13
Participation	4	4	3			11
Occupational performance	7	1				8
Technical aids	2	1	2			5
Remaining articles	32	13	8	2	1	56
Total	130	39	47	40	1	257

Discussion

The overall aim of this study was to review the research characteristics of articles with older persons as participants, aged over 65 years, with or without illness, disease or disabilities, reported in occupational therapy journals during the period 2013–2017. In this study, the proportion of articles focusing on older people in the selected journals was 8.7%. When Larsson et al. [8] published their study a decade ago (covering 2001–2006), the proportion was 15.5%. More journals (n = 15) and articles (n = 257) were included in this review compared to that review (journals n = 6, articles n = 212) [7], but as the total number of published articles in the selected journals is higher (N = 2960/N = 1368) [8], the proportion has decreased. It is noteworthy that the proportion of articles published within occupational therapy journals with older persons as participants has decreased precisely at a time when the fastest growing age group is those over 60 years [1]. This is an age group that will require a variety occupational therapy services [2] based on credible evidence. Thus, there is a need to improve research in the field of occupational therapy about older people, in relation to both quantity and category of research.

A recent bibliometric analysis of occupational therapy publications in general [18] presented the top five most productive countries as the United States of America, Australia, Canada, the United Kingdom, and Sweden. This differs from the results of the present study, where the top five most productive countries were the United States of America, Australia, Sweden, Canada, and Japan. Furthermore, there is a larger total numbers of articles for all age groups published by scholars in the United Kingdom and Canada in comparison to Japan and Sweden [18]; therefore, it is reasonable to assume that, in Sweden and Japan, research with older people as respondents is a prioritized research area. This bibliographic analysis [18] covers the years 1991–2014 and was conducted using a different research method, which might explain this difference.

The most common way to describe the group presented in the articles of interest for this review was to use a diagnosis, such as stroke or dementia. Many articles in the review used unclear groupings to describe the participants, such as older persons, community-dwelling, older drivers, or patients. Methodological challenges due to unclear groups have been reported previously in occupational therapy research, with consequences such as non-representative samples and decreased internal validity amongst the recruited sample [24]. The use of unclear labels, such as ‘older people’ or ‘community-dwelling’, could thus make it difficult to translate the results into practice because there could be something more at play than just the age or allocation of the participant. The importance of transferability and generalization has to be considered in relation to how a group is defined. Thus, there is a need for more precise presentations of the groups who are subject to research in future publications. An unclear group, such as community-dwelling, could include persons with many different health issues and diagnoses that may require different occupational therapy services. Furthermore, in some of the articles there was a huge age span among the participants. A wide age range amongst participants may affect the study’s internal validity because older and younger people may respond differently towards an intervention [25].

The numbers of included participants in the qualitative articles ranges from 1–149 participants, with a median of 10. Potter and Wetherell [26] argue that a sample size of around 10 participants is suitable when conducting qualitative research. If the sample is larger, it might be difficult to conduct an in-depth analysis of the data corpus [27]. The numbers of included participants in the quantitative articles ranges from 1–27,131 participants with a median of 61. A sample size consisting of less than 30 participants is considered to be relatively small, which will affect, among other things, the ability to generalize the results [28]. In general, the findings suggest that both qualitative and quantitative articles published in occupational therapy journals seem to use appropriate sample sizes. However, the lower quartile for both research designs (qualitative studies Q₁ = 6; quantitative studies Q₁ = 28) in relation to sample size is small. Furthermore, quantitative effectiveness studies have a small sample size (Q₁ = 11; median = 21). Effectiveness studies support practice [23], and a majority of these articles were investigating interventions. Since these effectiveness studies were conducted with small sample sizes, they can only be regarded as pilot or feasibility studies. In order to support practice, future effectiveness studies with larger sample sizes are needed.

The quality and design of available research are vital factors when deciding whether findings are valid and clinically useful for practitioners in their everyday work [5]. In this review, we found that the majority of research involving older people was conducted with a quantitative design using instruments for data collection. The quantitative research reported in the journals is spread between basic, correlational or descriptive, effectiveness, and instrument-testing research. This is consistent with previous findings [9]. The qualitative research published in occupational therapy journals has a limited spread in the different categories of research. In order to support occupational therapists in their everyday work with clients, both research aims and questions of a qualitative character should be developed to address other categories of research as well. For example, research questions in the category correlational and descriptive studies (i.e. articles demonstrating a link between occupational engagement and health) [23] could address how a client experiences an intervention in relation to their progress in rehabilitation. The client’s perspective on an intervention in evidence-based practice is of equal importance to the best available research and the clinical experience of the occupational therapist [4].

Qualitative research is mainly conducted using interviews as the data collection method. To address other categories of research, both research aims and questions could be developed to encourage different data-collection methods. For example, in the research category of instrument testing [23], video or audio recordings from settings and situations where occupational therapists actually use instruments with older clients could be used to study what occurs in these situations. There are closely related fields of research using naturalistic data by which researchers within the field of occupational therapy could be inspired (see, for example [29]).

Study limitations

This study does not attempt to examine all the gerontological research in occupational therapy but rather strives to review research involving older people published in international occupational therapy journals. Thus, the inclusion of articles was not based on a literature search in different databases. In this review, the inclusion of articles was conducted by examining the chosen journals for relevant articles. Similar search strategies have been used by other researchers to identify and analyze the content of articles in one [9,12,13] or more occupational therapy journals [8,11] in order to synthesize the areas, topics or characteristics of available research. We acknowledge that this search strategy does not cover all the available research published regarding older people within the field of occupational therapy that researchers publish in high-ranked interdisciplinary journals [20]. However, covering all available research was outside the scope of this article.

The articles included in this study were not graded in relation to evidence. In further studies, this is something that could be done in order to add another dimension to the findings. If such a grading is to be undertaken in future similar reviews, it is important to use an evidence grading system that supports different types of research aims and methodologies.

Disclosure statement

The authors report no conflicts of interest, and are responsible for the content and writing of the article.