Changes in health parameters in older lay volunteers who delivered a lifestyle-based program to frail older people at home

The volunteers, so-called buddies, were recruited with the help of a local volunteer organization providing care to old people (“Wiener Hilfswerk”). Participants answered advertising campaigns published in newspapers as well as informative articles in brochures. Furthermore, the program was presented in several television programs. Inclusion criteria were: older than 50 years of age, readiness to participate in both the control and the intervention group and willingness to participate for the full duration of the program intervention (6 months). Before the buddies started with the home visits, they received 4 training sessions over 4 weeks (each about 3 h long). These training sessions aimed to provide the buddies with the knowledge and skills to conduct home visits. After the training session, buddies were matched with a prefrail or frail partner, and the couple randomly assigned in an intervention or a control group. With prefrail or frail individuals in the intervention group, the buddies had to conduct six strength exercises and had to cover nutritional topics. With prefrail or frail individuals in the control group they were to conduct home visits with social support for the first 12 weeks. Afterwards they performed the same physical training and nutritional intervention. The training sessions covered topics of aging, malnutrition, importance of healthy nutrition and physical exercise, basics of physical training, as well as motivational techniques. Importance of appropriate execution of strength exercises until muscular exhaustion was carried out in practical sessions, where they also received instructions on how to teach the techniques. These were held by professionals (medical doctor, nutritionist, nutrition scientists, sports scientist, physiotherapist, and psychologist) and written materials designed for the study were given out to the participating buddies. All buddies were also trained concerning healthy nutrition. In this context buddies got information about topics such as proper hydration, protein and energy intake. Additionally, themes such as fruit and vegetables intake and “extras” (e.g., food with high fat, high salt and high sugar content) were addressed. As the control group, started with the nutritional and physical exercises 12 weeks after the intervention group, an additional refresher session was given to the buddies allocated to the control group.

During the home visits, buddies were asked to perform the same 6 strength exercises as prefrail individuals in 2 sets with 15 repetitions until exhaustion. Those were mini squats in front of a chair, chest presses against elastic resistance, an exercise for the abdominal muscles (performed sitting on a chair), hip extensions in standing position, reverse butterfly and shoulder presses against elastic resistance. If the described exercise was too easy, buddies were also provided with an alternative for each exercise, aiming to train the same muscle group. Apart from exercising during the home visits, buddies were asked to perform the same exercises once a week at home (without the frail persons), irrespective of the group assignment. Additionally, as described, they were trained concerning healthy eating in later adulthood and were asked to implement these nutritional guidelines in their everyday life.

Measurements in the buddies were conducted 3 times during the study duration: at baseline, after the first 12 weeks (visit 1) and after 24 weeks (visit 2). These were done by physical measurements of muscle strength and body impedance analysis as well as a number of questionnaires. Handgrip strength was assessed in kilograms (kg) using a hydraulic hand dynamometer from Jamar (Lafayette Instrument, Lafayette, Louisiana, USA), while the participants were sitting [21]. The strength of each hand was alternately tested three times, with a break of 1 min in between. The highest of all six values was taken for the calculation. The one repetition maximum (1-RPM) and muscular endurance were measured by using the “Concept2 DYNO” (Concept 2, Morrisville, Vermont, USA) dynamometer. One repetition maximum and muscular endurance were measured for three different exercises with their respective muscle groups. These included the seated squat (for lower limb muscles), bench press (pectoral muscles), and bench pull (upper back muscles). Using the “Concept2 DYNO” which uses a flywheel to create resistance ranging from 0 to 500 kg, dynamic measurements of the 1‑RPM and muscular endurance could be undertaken. A total of three submaximal warm-up lifts were done before the participants were instructed to do one repetition with maximum force possible (which is equal to their 1‑RPM). For the muscular endurance 15 consecutive repetitions with maximum voluntary effort were performed with the mean weight as the indicator of muscular endurance [22].

Anthropometry: body height was measured with a tape, and body weight with a Marsden MS-4203 (Marsden Group, Rotherham, UK) calibrated scale allowing body mass index (BMI) to be calculated. Abdominal girth was measured by a tape at a level 2.5 cm vertically higher than the umbilicus in expiration [23].

Body composition was measured by bioelectrical impedance analysis (BIA) in a standardized procedure involving the participants lying on their back with electrodes put on the dominant hand and foot. By using an alternating current body resistance and reactance are assessed. The BIA 2000-S device from Data Input (Darmstadt, Germany) was used in our study [24]. For our study; phase angle (PA), lean body mass and fat mass index were variables chosen. Although the pathological effects of the PA have not been clarified it has been interpreted as a sign of cell death or increased permeability of the cell membrane and is also been found to be a good prognostic marker of disease progression and survival. It was also reported that the PA increases with improvements in the clinical health status [25, 26]. The fat mass index was found to be a good indicator in assessment of body adiposity and is calculated as the body fat mass value of the BIA divided by the height squared [27]. Finally, the lean body mass in kg was calculated by using the formula “total body water/0.73” [28].

Physical Activity Scale for the Elderly (PASE): is a 12-question self-report questionnaire that measures physical activity in individuals aged 65 years and over in items that are organized regarding occupational, leisure and household activities during the last 7‑day period. Items are differently scored where leisure and strengthening activities are scored as never, seldom (1–2 days per week), sometimes (3–4 days per week) and often (5–7 days per week). Duration is then scored as less than 1 h, 1–2 h, 2–4 h or more than 4 h. Household and work-related items are scored as “yes” and “no”, where in work-related activities paid and unpaid work are again scored on the basis of duration. Overall score ranges from 0 to 400 with higher scores indicating more physical activity [29].

Mediterranean Diet Adherence Screener (MEDIAS): is a 14-item screener questionnaire consisting of questions about food amount intake and adherence to a Mediterranean diet. The answers are scored from 0–1 making the possible final score as anywhere between 0 and 14 points [30].

Protein intake was assessed by using the protein-containing food groups of the European Prospective Investigation into Cancer and Nutrition Study Food-Frequency Questionnaire (EPIC FFQ; [31]). The participants indicated the number of times a given food item was consumed, where one portion size corresponds for example to one palm of the hand, which was presented by an image in the questionnaire. A total of 39 food items which corresponds to 5 food groups as “meat and meat products and eggs” (9 items), “fish” (8 items), “milk and milk products” (7 items), “legumes and nuts” (6 items) and “cereals and bread” (9 items) were assessed. Each food item of the FFQ represents an individual food (for example salmon) and one question about the frequency and type of commonly consumed dietary supplements was added to the FFQ. Furthermore, the frequencies of food items were calculated as the daily intake. The German Nutrient Database “Bundeslebensmittelschlüssel” (BLS, version 3.01) was used to obtain the protein intake of one portion size (g per day). The daily protein intake was further calculated in g protein per kg bodyweight per day (g/kg BW/day). Moreover, the dietary protein intake was categorized in plant and animal-based protein intake (g/kg BW/day). Additionally, serum albumin concentration (analyzed by a licenced biomedical laboratory) was used to assess the nutritional status.

World Health Organization Quality of Life Scale Brief Version (WHO-QOL-Bref): a shorter version developed from the original WHO-QOL-100 measures quality of life in 4 domains: physical, psychological, social and environmental and a fifth overall QoL and general health domain were also calculated. Questions are on a Likert type scale (ranging from 1 as “completely disagree” to 5 “completely agree”) where the participants show their agreement or disagreement with a given statement. Final score is a sum of points, reverse scoring is applied where appropriate [32].

Social Support Questionnaire (Fragebogen zur sozialen Unterstützung; F‑Sozu): for this study, the short 14-item form was used, which was derived from the original 54-item version, and is recommended for research with limited time, but with validity and reliability remaining at a high level. The questionnaire consists of 14 statements, where on a Likert type scale the participants indicate their agreement or disagreement ranging from 1 “fully applies” to 5 “does not apply” [33]. Based on the median split and the percentiles the scores are divided into categories based on the level of social support needed.

As buddies of the intervention and also the control group were asked to implement the physical training exercises and the nutritional topics in their everyday life from the very beginning of the study, they were considered as one group and were analyzed as such. Descriptive statistics were done for each variable and presented as means, standard deviation in metric variables and frequencies in categorical variables. The relationship between handgrip strength and leg press muscle strength was investigated using Pearson’s product-moment correlation coefficient. Changes over time for relevant variables were calculated by the paired samples t‑tests and the Wilcoxson-Mann-Whitney test for non-parametric distribution in metric variables. Differences in categorical variables were calculated with the χ²-test. For all calculations, a statistical probability of p < 0.05 was considered significant. Statistical software package SPSS 20.0 (SPSS, Chicago, IL, USA) was used.