Exploring changes in dietary intake, physical activity and body weight during chemotherapy in women with breast cancer: A Mixed‐Methods Study

Quantitative data collection

From the Appetite, Hunger, Sensory Perception (AHSP) questionnaire, ²⁷ a 29‐question self‐assessment tool answered on a five‐point Likert scale (Table 1), we used three categories: taste (eight items, score range 8–40), smell (six items, range 6–30) and appetite (six items, range 6–30). Higher scores indicate a more positive self‐judgement on taste, smell and appetite. Trained dietitians conducted two 24‐h dietary recalls by telephone on randomly chosen days at T2, using a standardised protocol. The intake of total energy (kcal) was calculated in the computation module of Compl‐eat™ using the Dutch Food Composition Table 2013. ²⁸ BW was assessed with a dual‐energy X‐ray absorptiometry (DXA) scan at T1 and T3. The Short Questionnaire to Assess Health‐enhancing physical activity (SQUASH) was used to assess PA. ²⁹ Reported activities were subdivided into three intensity categories (min day^–1): light [1.6–2.9 metabolic equivalent of tasks (MET)], moderate (3–5.9 METs), vigorous (≥6 METs). One MET is the energy expenditure at rest. We then calculated adherence to the 2017 Dutch Physical Activity guideline, ³⁰ which recommends a minimum of 150 min of moderate to vigorous activity per week. The Multidimensional Fatigue Inventory (MFI‐20) is a self‐reporting tool consisting of twenty propositions on five dimensions of fatigue (general fatigue, physical fatigue, reduction in activity, reduction in motivation and cognitive fatigue) and its consequences. ³¹ , ³² , ³³

Qualitative data collection

The semi‐structured interviews were based on an interview guide that included topics such as side‐effects of chemotherapy (focus on smell and taste), experiences and explanations of changes in DI, PA and BW, as well as perceptions on dealing with these changes. ²⁶ The interviews were conducted at the participant's home. Oral permission for audiotaping was obtained before each interview.

Quantitative data

We first dichotomised the women based on EI at T2, PA and BW at baseline (T1) and T3. All intake parameters were complete for 117 participants of the total COBRA population (n = 181).

We compared EI during chemotherapy (T2) with the mean (SD) EI of 1779 (55.7) kcal day^–1 of the total COBRA population of n = 117 persons during chemotherapy (12) and created two groups: above (high) or below (low) that averaged energy intake. For PA at T1 and T3, we applied the Dutch Physical Activity Guideline ³⁰ ; performance below the guideline at T3 compared to T1 was classified as inactive and above the guideline at T3 compared to T1 was classified as active. For changes in BW, we calculated the changes in kg between T1 and T3 measured by DXA.

In addition, we compared taste, smell and appetite with the mean scores of the total COBRA population ¹² : for taste [mean (SD) 22.0 (0.57)], smell [mean (SD) 20.6 (0.42)] and appetite [mean (SD) 18.7 (0.50)]; above (high) or below (low) mean scores. For fatigue at T1 and T3, we calculated the MFI, and compared the values with the mean scores of a reference group of cancer survivors; above (high) or below (low) the mean scores. BMI was classified such that <25 kg m^–2 was considered low and BMI ≥ 25 kg m^–2 was considered high.

Qualitative data

All audiotaped interviews were transcribed verbatim and a synopsis was written for each respondent. To identify, analyse and describe patterns in respondents’ individual experiences, athematic analysis ³⁴ , ³⁵ was carried out during which, after close reading and coding of the transcribed interviews, we formulated subthemes and eventually overarching themes. Interview data were managed with maxqda (VERBI Software, Marburg, Germany. ³⁶ All subthemes and themes were described in a mind map: one for each individual. In this way, we identified the essence of each theme, searched for relations through constant comparison across cases, looked for deviant cases, and analysed variation within and between cases.

Combining quantitative and qualitative data

All mind maps included individual results of the quantitative measurements (AHSP, BW, 24 h dietary recall, SQUASH and MFI) to find characteristics that further differentiate the themes. Subsequently, we gathered the relevant (sub)themes and quantitative measurements necessary to answer the research question. Finally, all the data were fit into the overarching story emerging from the data, aiming to gain insight into differences and similarities between the women regarding the findings from the quantitative and qualitative analyses.

Dietary intake

The average reported EI of the 25 women during chemotherapy of 1761 kcal (range 1182–3102 kcal) was similar to the average reported EI of all women in the total COBRA study of 1779 (55.7) kcal. ¹² Women (n = 11) with an EI above average (mean 2165 kcal; range 1783–3102 kcal) were almost equally distributed between a lower BMI (n = 6) and a higher BMI (n = 5), just like women (n = 14) with an EI below average (mean 1442 kcal; range 1182–1738 kcal) were equally distributed between a higher BMI (n = 7) and a lower BMI (n = 7). Women with an EI above average were approximately as active (60% versus 54% met PA recommendation) and as fatigued (70% versus 77%) as women with an EI below average during chemotherapy. Those with a higher EI less often gained BW (64% versus 79%). Scores on appetite, taste and smell were low for most women, regardless of their EI. Women with lower EI during chemotherapy more often had a lower taste score (91%) than women with a higher EI (70%) (Table 3).

Physical activity

All women were moderately intensely physically active for ≥150 min per week at T1. At T3, approximately half of the women (13/23) were physically active for more than 150 min per week. Active women during chemotherapy (women who were active at T1 and stayed active at T3) (n = 13) more often had a lower BMI (69%) before treatment than inactive women (women who were active at T1 and below guideline PA at T3) (n = 10) (30%). They were more likely to have an EI above average (60% vs. 40%). The majority gained BW (62%), although inactive women were more likely to gain BW (90%). Active women during chemotherapy were less fatigued (54% versus 100%) and less often showed appetite problems (100% versus 58%) and smell problems (100% versus 50%) compared to inactive women (Table 3).

Body weight

Women had a mean baseline (T1) BMI of 24.9 kg m^–2 (19.3–32.4). Their average BW at T1 was 71.4 kg (range 54.0–93.1 kg), the average difference in BW between T1 and T3 was 0.8 kg (range −6.5 to +4.5 kg). The majority of women (72%, n = 18) showed an increase in BW between T1 and T3: mean +1.9 kg (range 0.1–4.5); 28% (n = 7) showed a decrease in BW: mean −2.1 kg (range –0.2 to −6.5 kg). Women who gained weight were less likely to have an EI above average (39%) than women who lost weight (57%). Women with weight gain were less likely to meet the PA recommendation during chemotherapy (47% vs. 83%) and more often had a low BMI at baseline (61% vs. 29%). Women who gained weight and women who lost weight both showed high levels of fatigue (76% vs. 67%), appetite problems (76% vs. 80%) and taste problems (82% vs. 80%). Women with weight gain reported less smell problems (88% vs. 100%) than women with weight loss (Table 3).

Dietary intake

Most women with an EI above average reported they perceived their diet was healthy before diagnosis and were able to maintain their diet despite chemotherapy‐related complaints (Table 4). Most women with low EI indicated they did not always eat healthily and were less aware of what healthy eating was before the diagnosis. These women continued to struggle during chemotherapy. Sometimes eating was such a problem that they were happy when they could eat at all, healthily or not:

… happy when it went down, no matter what it was. Preferred sausage rolls to bread, baked potatoes and fries were tasty, no more fight against candy. [woman with low EI, BMI high, PA low, fatigue high, BW increased, Taste (T), Smell (S) and Appetite (A) low]

Some women struggled because of extremely bad taste, especially during the last three courses with Docetaxel, and could not eat at all.

Some women with higher intake were matter‐of‐fact about their eating problems:

The chemo is temporary, I wasn't looking for a solution; just keep eating as normally as possible, even if it's not tasty … [woman with high EI, BMI high, PA not available (NA), fatigue NA, BW decreased, T, S, A: NA]

Women with lower intake indicated that they first needed to experience what it was like to be treated with chemotherapy. They found it difficult when health care professionals (HCPs) recommended ‘a normal diet’ during chemotherapy, because they assumed that a normal diet was not good enough and they longed to do more:

I was searching for supplements, vitamins yes or no, no grapefruit, but orange? What is or isn't healthy to eat now? (woman with low EI, BMI high, PA low, fatigue high, BW increased, T, S and A low)

Women with higher EI looked for creative solutions to be able to continue their intake, such as adding herbs, ketchup, salt for better taste; sucking ice cubes to avoid a metallic taste; eating yoghurt instead of bread; eating fruits such as grated apple because it went down easier. They mainly cooked easy‐to‐prepare meals, such as oatmeal in the morning and ready‐to‐eat meals in the evening. They more often ate smaller portions throughout the day. Some women reported that they perceived eating more healthily than before diagnosis and added various healthy alternatives and supplements to their diet.

Women with lower EI also tried to change their eating habits to accommodate a damaged and dry mouth, no appetite and loss of taste, although their aversion to almost every food was hard to overcome. They continued to struggle because they realised that it was necessary and called it functional eating:

I eat because I know I have to, but it's not going well and I certainly don't enjoy it (woman with low EI, BMI high, PA low, fatigue high, BW increased, T, S and A low)