Heart girth best predicts live weights of market-age pigs in Tanzania Authors: Mwemezi L. Kabululu, Anselme Shyaka, Anselme Shyaka, Anselme Shyaka Published: 2023 ABSTRACT The aim of this study was to use linear body measurements to develop and validate a regression-based model for prediction of live weights (LW) of pigs reared under smallholder settings in rural areas in the southern highlands of Tanzania. LW of 400 pigs (range 7 to 91 kg) was measured, along with their heart girths (HG) and body lengths (BL). BL was measured from the midpoint between the ears to the tail base. HG was measured as chest circumference just behind the front legs. LW was determined using a portable hanging scale. An analysis of covariance was performed to test for differences in LW between male and female pigs, including age, HG and BL as covariates. LW was regressed on HG and BL using simple and multiple linear regressions. Models were developed for all pig ages, and separately for market/breeding-age pigs and those below market/breeding age. Model validation was done using a split-samples approach, followed by PRESS-related statistics. Model efficiency and accuracy were assessed using the coefficient of determination, R2, and standard deviation of the random error, respectively. Model stability was determined by assessing ‘shrinkage’ of R2 value. Results showed that HG was the best predictor of LW in market/breeding-age pigs (model equation: LW = 1.22HG—52.384; R2 = 0.94, error = 3.7). BL, age and sex of pigs did not influence LW estimates. It is expected that LW estimation tools will be developed to enable more accurate estimation of LW in the pig value chain in the area. 1. Introduction In Tanzania, pig/pork production and marketing have continued to record a remarkable growth over recent decades, as compared to other livestock sectors, contributing substantially to improving livelihoods of rural and urban communities [1, 2]. Most (>90%) of the pigs in Tanzania are produced in low-input output systems, largely in rural areas, involving smallholder backyard units with herd sizes of usually less than 10 pigs [1]. Pig marketing in many areas of Tanzania largely involves farm-gate transactions between a pig trader and a farmer [3]. Traders usually buy pigs based on their subjective visual estimates of weight, in most cases to their advantage by underestimation (Walugembe et al., 2014) [4]. This results in farmers receiving less money than they would have rightfully deserved. This could be addressed by valuing and selling/buying pigs on the basis of price/kg of live weight (LW). Apart from its usefulness in pig marketing, estimation of pig weight is important for optimization of management, performance and hence profitability [5, 6]. It enables accurate calculation of feed requirements [7, 8]; monitoring of pig growth and health status; and determination of correct drug dosages [9, 10]. Therefore, the importance of having optimally accurate and affordable methods for estimating pig weights cannot be overemphasized. There are two methods which can be used to estimate pig weights, direct and indirect methods [11]. A direct method involves the use of calibrated electronic or mechanical weigh scales, while indirect methods include visual estimation of weights; imaging and use of linear body measurements. Despite being an accurate and widely accepted method globally, the use of weigh scales is limited in most smallholder pig farming settings, Tanzania inclusive. Apart from inaccessibility, their use requires high input of time and labour, and can cause stress and injury to pigs and handlers [12, 13]. Another drawback is the lack of technical skills in operating and maintaining the scales [14, 15]. Due to the limitations, indirect methods of weight estimation present a better option, with the use of body measurements being the most common [11]. Body measurements provide a relatively accurate, consistent and cheaper method of estimating LW. Several studies in different countries have described the relationship between the linear body measurements and pig weight, and developed weight prediction models [4, 12, 16–18]. However, due to differences in management practices, and in pig body sizes and conformations, weight estimation methods developed in one locality may not be suitable for use in another locality [6, 19]. Hence, for optimal accuracy, weight estimation methods should only be applied to pigs reared under similar conditions of management, and more or less of the same genotype. Therefore, this study had two objectives. Firstly, the study aimed to develop and validate a regression-based equation which can be used to accurately estimate LW of pigs reared by smallholder farmers in rural areas of the southern highlands of Tanzania, by using simple linear body measurements, in this case, heart girth (HG) and body length (BL). In addition, this study also aimed to validate use of a commercially available calibrated heart girth tape for use on market-age pigs in the area, by assessing difference in mean between actual weight measured using a hanging scale and weight estimated by the heart girth tape. 2. Materials and methods 2.1 Statement of ethical consideration The conduct for this study fitted within the conduct of the field intervention trial which was approved by the Research, Publications and Ethics committee of the College of Veterinary Medicine and Biomedical Sciences (CVMBS) of the Sokoine University of Agriculture, SUA, in Tanzania (Reference number: SUA/CVMS/016/32). All participating farmers provided written informed consent to participate. The trial was conducted as guided by the principles of good clinical practice. Animal handling adhered to Animal welfare regulation stipulated in section 40 of the Tanzania’s Animal Welfare Act number 19 of 2008. Animals were handled by livestock technicians who also received pre-trial training on best practices in handling animals. 2.2 Study sites and study animals The study was conducted in 10 villages from Mbeya rural and Mbozi districts located in the southern highlands zone, south western Tanzania. The two districts are, to a great extent, rural and pig rearing was largely on a small scale, with majority of farmers keeping one to five pigs [2]. Most of the pigs were nondescript, being crossbreeds of indigenous breeds and exotic breeds mainly Landrace and Large white [20, 21]. Pig rearing practices were rudimentary. A study by Kabululu et al. [21] showed that about 60% of pigs are kept in pens while 20% are left to scavenge for food. Constructed pens had either raised floors (slatted with timber off cuts or with tree logs/sticks) or ground floor (earthed or with concrete). The pen walls had either timber off cuts, tree poles/sticks, bamboo poles or bricks. Roofing materials used were grass/thatches, bamboo poles or iron sheets. Maize bran and kitchen wastes were the main feeds provided to pigs with little or no protein or mineral supplementation. Selling/buying of pigs relied solely on visual estimates of body size/weight of pigs with virtually no actual weighing of pigs. 2.3 Data collection Data were collected during household visits in June to July 2017. Weights and body measurements were recorded from 400 pigs, excluding sick, pregnant and nursing sows. From the list of all households whose heads had consented to participate in the main (intervention) study [22], pig measurements (LW, HG and BL) were taken from a maximum of two pigs, in every other household that was visited. To ensure accurate measurements, pigs were restrained and their heads stabilized by using a pig snare. HG and BL were measured by using ANIMETER®, a calibrated commercial weigh tape (BA 219 Animeter 050, Albert Kerbl GmbH, Buchbach, Germany). BL, measured on the back of the animal, was defined as the length (in centimeters) from the midpoint between the ears to the base of the tail [4]. HG, also in centimetres, was defined as the chest circumference, measured by wrapping the measuring tape snuggly around the chest area, just behind the front legs. Pig weighing was done using a portable hanging scale (with a precision of 500 g). Pig sex and age (in months) were also recorded. Estimation pig age was done by pig owners, and by confirmation by using available records. For validation of a HG tape (ANIMETER®), LWs of a random sample of 40 pigs (from a sampling frame of the 400 pigs) were estimated using the tape, in addition to weights measured by the hanging scale. This was achieved by including every 10th pig, from all the pigs that were measured. The tape had values of estimates of LW printed on it, corresponding to measured HG values. 2.4 Data analysis Analysis of data was done using STATA© (Stata Corporation, College Station, Texas) version 14. Descriptive statistics were used to determine means (± SD), medians, minimum and maximum values, and 25% and 75% percentiles of different parameters. Pearson pairwise correlation analysis was performed to determine the degree of a linear relationship between body measurements and LW. A preliminary analysis of covariance (ANCOVA) was performed to test for differences in LW between males and females; including age, HG and BL as covariates included as main effects and as interaction terms of Age*HG and Age*BL. LW was regressed on HG, BL and age using simple and multiple linear regressions. The general linear equation used was; Y=β0+βiXi+ε, where; Y = Live (body) weight β0 = constant (intercept) βi = regression coefficient of the ith independent variable (slope of the regression line) Xi = The value of the ith independent variable ε = residual (error term) Regression analysis was performed in two stages, first by including all pigs and second by categorizing pigs into those below 12 months (below market/breeding age), and those 12 months and above (market/breeding-age pigs). In the first stage, a multiple regression analysis was performed regressing LW on HG and age as main and interaction effects; and simple regression analyses regressing LW on each of the two variables (HG and age) separately. In the second stage, a simple regression of LW on HG was performed for each of the two age categories. The resulting regression models were tested for normality and homoscedasticity of residuals, both visually and computationally. Normality was tested using a Shapiro-Wilk test. Homoscedasticity, an assumption of constant variances of the residuals was determined by assessing scatter plots of residuals against fitted values and using Breusch-Pagan/Cook-Weisberg test. Model efficiency was tested using the coefficient of determination (goodness of fit, R2), the percentage of the variance in the LW explained by predictors. Model accuracy was estimated by assessing root mean square error (RMSE), which is the standard deviation of the random error. Further, the accuracy of the more predictive model was determined from residual statistics. To explain distribution of the residuals, mean, median, 25% and 75% percentiles were determined. Mean absolute value of residuals was used to determine average prediction error. Prediction intervals and confidence intervals around mean predicted weight were determined using lower and upper bounds of predicted values and standard errors of forecast and prediction, respectively. For model cross-validation, a split-samples approach was used, followed by PRESS-related statistics. A random selection of 75% of all pigs (model derivation group) was used to develop the model and the remaining 25% (model validation group) was used for model validation [18, 23, 24]. PRESS-related statistics involved the use of PRESS, which is the predicted residual sum of squares. The more predictive model was used to estimate weights of the pigs in the validation group, and PRESS was determined from the calculated residuals. Then the PRESS was used to calculate a R2 PRESS, a modified form of R2, using the formula: R2 PRESS = 1 –[PRESS /SS total], where SS total equals the sum of squares for the original regression equation [25, 26]. Model stability (hence reliability) was then assessed by estimating ‘shrinkage’, which was the difference between R2 and R2 PRESS. The smaller the ‘shrinkage’, the more stable the model was considered to be. The difference in mean between actual weight measured by the scale and weight estimated by the heart girth tape was compared using a paired t test. Differences between measured weights and weights estimated using the tape were considered as residuals, and these were compared to the differences between measured weights and weights predicted by the model. Difference in mean absolute residuals, together with 25% and 75% percentiles were determined. 3. Results Among the 400 pigs included in the study, 242 (60.8%) were females and 156 (39.2%) were males. Age, LW, HG and BL all measured pigs are reported in Table 1. Two pigs had missing age information. Age ranged from four to 44 months, with a mean of 10 months and a median of 8 months. When categorized into age groups, 299 (74.7%) were those below market/breeding age (below 12 months) and 101 (25.3%) were market/breeding-age pigs. Mean live weight was 28.4 kg, with a mean of 24.3 kg for pigs below market age and 40.3 kg for market/breeding-age pigs. 10.1371/journal.pone.0295433.t001Table 1Means, standard deviations (SD), minimum, maximum, range of live weights, age, and body measurements (heart girth and body length) of 398 pigs (242 females, 156 males) pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania.ParameterSex*MeanSDMinimumMaximumAge (months)Males8.63.7432Females10.95.6444All pigs10.15.1444Live weight (kg)Males25.29.6860Females30.513.7791All pigs28.412.5791Heart girth (cm)Males62.79.33891Females67.811.941118All pigs65.711.338118Body length (cm)Males73.812.145103Females80.114.543137All pigs77.61443137* sex of two pigs was not recorded Means, standard deviations (SD), minimum, maximum, range of live weights, age, and body measurements (heart girth and body length) of 398 pigs (242 females, 156 males) pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania. * sex of two pigs was not recorded Results of a pairwise correlation analysis (Table 2) showed that, overall, HG had the highest correlation with LW (r = 93%) while age and LW were moderately correlated (r = 59%). Age was also moderately correlated with HG and BL. Stratified by age groups, higher correlations were found in pigs aged 12 months and above. 10.1371/journal.pone.0295433.t002Table 2Pearson correlation coefficients (r) for correlation between live weight and body measurements (heart girth and body length) of 400 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania.ParameterLive weightHeart girthBody lengthAge All pigs (n = 400) Live weight----Heart girth0.93*---Body length0.860.89--Age0.590.560.54- Pigs below market/breeding age (n = 299) Live weight----Heart girth0.89---Body length0.790.84-- Market/breeding-age pigs (n = 101) Live weight----Heart girth0.96---Body length0.860.86--*All correlations were significant (p = 0.000) Pearson correlation coefficients (r) for correlation between live weight and body measurements (heart girth and body length) of 400 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania. *All correlations were significant (p = 0.000) The analysis of variance (Table 3) showed that pig sex had no effect on LW (p = 0.571). HG showed a significant effect on LW as both main and interaction effects were significant. Age also had a significant effect on LW, both as a main effect and as an interaction with HG. However, BL had no significant relationship with LW, hence it was not included in the subsequent regression modelling. 10.1371/journal.pone.0295433.t003Table 3Analysis of variance of the difference in live weights between male and female pigs including age, heart girth and body length as covariates.Data were collected from 400 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania.SourcePartial SSdfMSF-valueP-value Model 55,398.969,233.2565.50.000 Sex 5.215.20.320.571 HG 861.31861.352.80.000 BL 17.6117.61.10.300 Age 676.11676.141.30.000 HG * Age 91.6191.65.60.018 BL * Age 11.7111.70.70.397*SS = sum of squares; df = degrees of freedom; MS = mean squares Analysis of variance of the difference in live weights between male and female pigs including age, heart girth and body length as covariates. Data were collected from 400 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania. *SS = sum of squares; df = degrees of freedom; MS = mean squares Results of the regression analysis (Table 4) showed that the model regressing LW on HG in market/breeding-age pigs (equation 5: LW = 1.222HG—53.384 + ε) was the best model. This means HG was best correlated with LW in market/breeding-age pigs (R2 = 93.8%) and prediction of LW from HG had the lowest prediction error (RMSE = 3.7). Age alone was not reliable in the prediction of LW (R2 = 32.9%). 10.1371/journal.pone.0295433.t004Table 4Results of the regression analysis of live body weight on heart girth and age of different categories of 400 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania.Pig groupParameterModel No.Regression equation R 2 adj. R2RMSE (ε)F value*All pigsHG, Age1LW = 0.989HG + 0.252BL − 39.237 ± ε86.486.34.7925.95HG2LW = 1.053HG − 40.931 ± ε85.685.64.81748.79Age3LW = 1.400Age + 15.062 ± ε32.932.610.4145.82Pigs below market/breeding ageHG4LW = 0.807HG − 26.931 ± ε74.974.84.2656.37Market/breeding-age pigsHG5LW = 1.222HG − 52.834 ± ε93.893.73.71108.79HG = heart girth; BL = body length; LW = live weight; R2 = coefficient of determination; adj. R2 = adjusted R2; RMSE = root mean square of error*all prediction equations were significant (p<0.05) Results of the regression analysis of live body weight on heart girth and age of different categories of 400 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania. HG = heart girth; BL = body length; LW = live weight; R2 = coefficient of determination; adj. R2 = adjusted R2; RMSE = root mean square of error *all prediction equations were significant (p<0.05) Homoscedasticity was verified for the model no. 5 (p = 0.39, X2 = 0.71), while it was not for the remaining models (equations 1–4). The normality test showed that residuals for the model equation no. 5 were normally distributed (p = 0.197, Z value = 0.853), while residuals for the remaining models (1 to 4) were not normally distributed. Fig 1 shows residual-versus-fitted plots following the regression analyses. In each of the plots A–D, corresponding to model equations 1–4, a pattern is observed, indicating violation of the least squares assumption. In the plots, an increase is observed in the variation of the residuals as the fitted values increase (heteroscedasticity). In addition, in plots A, B and D, the residuals show a curvature, indicating a violation of the assumption of linearity; while for plot C, there is skewing of the values of data points. On the other hand, plot E which corresponds to model equation number 5, shows a random and uniform scatter of points (residuals), indicating a good fit. Plots of residuals (Y axis) versus fitted values (X axis) following regression of live weight (LW) on heart girth (HG) and age of 400 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts, Tanzania. A = regression of LW on HG and age (pigs of all ages); B = regression of LW on HG (pigs of all ages); C = regression of LW on age (pigs of all ages); D = regression of LW on HG (pigs below 12 months); E = regression of LW on HG (pigs 12 months and above). Fig 2 shows line of best fit for the more predictive model, together with 95% confidence and prediction intervals. The confidence interval (shaded area around the best-fit line) shows the area where there is a 95% probability that true best-fit line lies. The prediction interval (whole area between the two outer lines) shows the 95% probability that weights predicted by the model will lie within this interval range. Scatter plot of live weight vs heart girth showing line of best fit, overlaid with 95% confidence interval (shaded area around the line of best fit) and prediction intervals (area between outer lines). “Shrinkage” was determined to be relatively low (0.031), and according to suggestion by Kleibaum et al. [27] (shrinkage <0.1 = stable model), the more predictive model (model 5) was considered to be stable and hence was validated. The regression model results indicated that weight increased by 1.22 kg as heart girth increased by 1 cm. The model yielded a prediction error of 3.7 kg while the mean predicted weight was 39.7 kg. This translates to 9.3% of the predicted LW. Mean residual was close to zero (-5.17e-09) indicating a good fit of the regression line. The model will predict weight with a mean error of ≤ 0.19 kg for half of the pigs and a mean error of ≤ 2.03 kg for 75% of the pigs. Paired t test showed that there was no significant difference in mean between weights measured by the hanging scale and weights estimated using the heart girth tape. Nevertheless, mean absolute residual was higher for weight estimation using the heart girth tape as compared to estimation using the model (Table 5). In addition, although the median values and 25% percentiles were comparable, 75% percentile was higher for residuals from weight estimation using the heart girth tape than for weight estimation using the model. 10.1371/journal.pone.0295433.t005Table 5Mean absolute differences between measured (live) pig weight and weight estimated by heart girth tape for a random sample of 40 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts.ParameterResiduals (weight predicted by the model)Residuals (weight estimated by the heart girth tape)Mean3.033.49Standard deviation2.492.80Median1.992.0025% percentile1.171.0075% percentile4.786.00 Mean absolute differences between measured (live) pig weight and weight estimated by heart girth tape for a random sample of 40 pigs reared by smallholder farmers in Mbeya rural and Mbozi districts. 4. Discussion In Tanzania, there is a paucity of studies which assessed correlation between body weight and linear body measurements of livestock. This study could be the first to determine the relationship between linear body measurements and body weights in Tanzanian pigs. The study has shown a strong correlation (R2 = 0.94) of HG and LW of market/breeding-age pigs in a major pig production area in the country. Hence, HG could be reliably used for estimation of LW of market/breeding-age pigs in the area, with relatively high accuracy. HG is probably the most commonly used body measurement parameter to predict LW in livestock. A number of studies have reported that HG is best correlated with LW in pigs, when used as a single predictor. In Ugandan pigs, HG has been reported to be the most predictive single body measurement, with an R2 value of up to 0.94 [4, 28]. In Zimbabwe, Sungirai et al. [19] reported a coefficient of determination of 0.89 for relationship between HG and LW in Large White and Landrace breeds. Machebe and Ezekwe [17] studied growing to finishing pigs in Nigeria and found that HG was highly correlated (R2 = 0.98) with LW. In Ghana, HG yielded the highest prediction ability (Adjusted R2 = 0.77 to 0.92) in large white pigs of five to 20 weeks [29]. In India, HG had the best correlation with LW of Landly pigs 9–15 months of age (R2 = 0.80) [13]. Studying Philippine native pigs weighing 20–50 kg, Paras et al. [30] found that HG explained 88% (i.e. R2 = 0.88) of the total variability in LW. In Ireland, O’Connell et al. [31] found that using HG alone, sow weight could be estimated with an (adjusted) R2 of 0.81. Groesbeck et al. [12] reported a strong correlation (R2 = 0.98) in grower to finisher pigs in the United States. Other studies had also reported a strong LW prediction ability when HG was combined with other body measurements. Mutua et al. [18] developed a single model combining HG and BL for predicting LW in pigs in rural western Kenya, and found that the model explained 88% to 91% of the total variation in LW for pigs from ≤ 5 months to breeding age (≥ 10 months). In Nigeria, Alenyoregue et al. [32] came up with a prediction equation with age of the pig, HG, and BL (R2 = 0.93). In the study by Kumari et al. [13], including BL and a quadratic term of HG led to R2 value of 82.9% compared to values of 58% and 80% when LW was separately regressed on BL and HG, respectively. Paras et al. [30] found that including BL as a predictor increased R2 from 88% to 92%. In this study, BL did not seem to significantly influence LW estimations, contrary to some previous studies. The reason for this could be the fact that HG largely measures muscle and fat tissues, which more closely reflect body condition (and hence LW) as opposed to BL which measures skeletal tissues [32]. The fact that HG could be used to predict LW with relatively high efficiency and accuracy makes it simple to develop LW estimation tools. In addition, HG has been considered to be relatively easier to measure under field conditions, as compared to other measurements [6]. HG has also been reported to be a significant predictor of LW in other types of livestock, such as cattle [15, 33, 34] and goats [6, 35, 36]. Previous studies have shown that age can be used to estimate LW of pigs [7, 19]. In this study, however, age was not found to influence LW. This can be explained by two reasons. Firstly, as the pigs in the area are mostly non-descript/ crossbred pigs, growth pattern can be expected to be different. Secondly, pig management practices in the area are generally rudimentary [21]. As a result, pigs cannot exhibit their full growth potential related to their age, and a sizeable number of pigs appeared smaller in size than their age. In addition, difference in pig management practices across households could be among reasons, as it was pointed out in Uganda [37, 38] A previous study in the area of the current study, by Kimbi et al. [3], reported that pig age was a poor determinant of price (R2 = 0.22) especially for older pigs. Although age itself was not a good predictor of LW, the observed significant interaction of HG and age indicates that age specific models could be more appropriate, as it was suggested by Mutua et al. [18]. This is more so considering the fact that HG was a significant predictor only in pigs one year and above. Further studies are suggested, including more body measurements/parameters in order to develop models for prediction of LW in younger pigs in the area. The results of this study have shown that sex did not influence LW estimation. This is in line with previous studies in Kenya [18], Uganda [4, 28] and Zimbabwe [19]. As pointed out before, the reason for this could be related to the pig management constraints which make the pigs unable to express their genetic potential. Under an ideal situation, accuracy of LW estimation depends on the purpose of weight estimation. The accuracy range found in this study (9.3%) can be considered to be acceptable for the purpose of determining treatment doses [39]. In addition, it is likely to lead to more reliable weight estimates during pig selling/buying compared to visual estimates. The heart girth tape assessed in this study was found to have a reasonable ability to estimate LW in market/breeding-age pigs in the area. It is assumed that the tape was developed for use with more exotic, well-managed pigs of Europe and North America [18]. Therefore, it was not expected to accurately estimate LW of pigs in the study area. Thus, the tape can reliably be used to estimate LW of pigs in the area until more efficient and accurate tools for weight estimation are developed. There are a number of ways the results of this study could be put to use. One way is to develop a measuring tape which has LW corresponding to HG values printed on it. Another approach is to develop a smart phone application where farmers can input HG values and readily get LW estimates. Lastly, HG and their corresponding LW can be produced in a paper-based chart which can be made available to farmers. 5. Conclusion In conclusion, this study has developed and validated a LW prediction model (LW = 1.22*2HG—52.834) with an efficiency of 93.8% and an average prediction error of 3.7 kg in predicting LW of market/breeding-age pigs reared by smallholder farmers in Mbeya rural and Mbozi districts in the southern highlands zone of Tanzania. It is expected the model will enable more accurate estimation of weight of market/breeding-age pigs in the area. In addition, this study has assessed and confirmed that the available commercially available heart girth tape had reasonable ability to estimate LW of market-age pigs in the area. Supporting information (XLSX) Click here for additional data file. 8 Aug 2023 PONE-D-23-16971Heart girth best predicts live weights of market-age pigs in TanzaniaPLOS ONE Dear Dr. Kabululu, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. ============================== ACADEMIC EDITOR: Dear Dr. Mwemezi L. Kabululu Thanks for submitting your manuscript for consideration by PLOS ONE. Two researchers reviewed your manuscript and provided comments and questions you need to consider in order to improve your manuscript. I look forward to receiving your rebuttal response. ============================== Please submit your revised manuscript by Sep 22 2023 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript:A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols. We look forward to receiving your revised manuscript. Kind regards, Anselme Shyaka, Ph.D Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability. Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized. Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access. We will update your Data Availability statement to reflect the information you provide in your cover letter. 3. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide. Additional Editor Comment: Dear Dr. Kabululu, Thanks for submitting your manuscript for consideration by PLOS ONE. Two researchers reviewed your manuscript and provided comments and questions you need to consider in order to improve your manuscript. I look forward to receiving your rebuttal response. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes ********** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: I Don't Know Reviewer #2: Yes ********** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: No Reviewer #2: No ********** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ********** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: This paper presents an analysis of pig measurements against weight, as a means to better estimate live pig weight int he absence of weigh scales. It is undertaken in Tanzania, and presents what appears as a thorough statistical analysis. This appears sound, but I would suggest that it be looked at by a statistician before the paper progresses. My main issue is that this is a crowded space of publication, amazingly. A paper not even cited by the author (https://link.springer.com/article/10.1007/s11250-023-03561-z) has effectively already done this study, coming up with a slightly different but entirely analagous result. It is unclear, other than having taken a slightly different approach, how this manuscript builds on that previously done work. Just being specific to Tanzania (and one small region within it) cannot be enough in this context - if the Marshall paper meets the need well, where is the novelty and the added value? On this basis I reject the paper, but the author may wish to respond with a thorough rebuttal. Reviewer #2: //journal to confirm if data has been made available (number 3 above) Line 38: delete -s in ‘involves’ Line 64 -68: it can be clarified that there were two objectives that the study sought to address (then show how each was addressed, and the results) Line 70: Good to mention that participants willingly consented to participate in the study (were other data collected, besides the measurements?) Line 93- 105: to make it easier to follow, perhaps have subheadings under 2.3 corresponding to the two objectives – i.e., development of the model, validation of existing weight estimation tape. What informed the sample size of 400? How were the 40 selected, it is mentioned randomly, but not clear how, in the field – was a list of pigs/ households made available before? I suggest we also have a summary description of these 40 Line 107: Can include the version used Table 1: may be instead of both use ‘all pigs’ Table 2: shouldn’t ‘all pigs’ be in the row below (n=400?)? please indicate, in brackets, the numbers for the other rows (pigs below market age, and market age pigs) Table 5: please clarify that this is only for the 40 pigs subjected to both weighing, and the tape—the model estimation results, is it for the 40 (not for all pigs)? References: review all, check number 18, some of the author names are missing ********** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: No Reviewer #2: No ********** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step. Submitted filename: reviewer comments June28.pdf Click here for additional data file. 28 Aug 2023 COMMENT My main issue is that this is a crowded space of publication, amazingly. A paper not even cited by the author (https://link.springer.com/article/10.1007/s11250-023-03561-z) has effectively already done this study, coming up with a slightly different but entirely analagous result. It is unclear, other than having taken a slightly different approach, how this manuscript builds on that previously done work. Just being specific to Tanzania (and one small region within it) cannot be enough in this context - if the Marshall paper meets the need well, where is the novelty and the added value? RESPONSE: The author appreciates the views of the reviewers. However, the following should be noted; -The very first draft of the paper was written during June to August 2022, way before the Marshall’s paper was published. At the time this paper was submitted to PLOS ONE, the author was not aware of the Marshall’s paper, hence could not have cited it. -Although this papers bears some similarities with Marshall’s, different approaches have been used. In addition, for this kind of studies, model developed in one area may not be suitable to be used in another area, due to inherent differences between pigs and management practices. Hence a model developed using Ugandan pigs may not be suitable for use in Tanzanian pigs. Hence, findings of this study are still deemed as important to inform tailor-made practice. A point to add is that, the area that the study was carried out is located within the major pig production area in Tanzania with about a half of the pig population; hence the results could more effectively represent the larger pig population COMMENT: Line 38: delete -s in ‘involves’ RESPONSE: Comment is received. However the ‘involves’ refers to the ‘pig marketing’, hence it is deemed as being right COMMENT: Line 64 -68: it can be clarified that there were two objectives that the study sought to address (then show how each was addressed, and the results) RESPONSE: Clarification has been made, as suggested COMMENT: Line 70: Good to mention that participants willingly consented to participate in the study (were other data collected, besides the measurements?) RESPONSE: A sentence has been added, to mention that farmers ’ willingly participated COMMENT: Line 93- 105: to make it easier to follow, perhaps have subheadings under 2.3 corresponding to the two objectives – i.e., development of the model, validation of existing weight estimation tape. What informed the sample size of 400? How were the 40 selected, it is mentioned randomly, but not clear how, in the field – was a list of pigs/ households made available before? I suggest we also have a summary description of these 40 RESPONSE: Description of how the 400 and 40 pigs were obtained (for model development and validation of the existing heart girth tape, respectively) has been improved to make it more clear COMMENT: Line 107: Can include the version used RESPONSE: A mention of STATA version used has been added COMMENT: Table 1: may be instead of both use ‘all pigs’ RESPONSE: Suggested correction has been made COMMENT: Table 2: shouldn’t ‘all pigs’ be in the row below (n=400?)? please indicate, in brackets, the numbers for the other rows (pigs below market age, and market age pigs) RESPONSE: Suggested correction has been made COMMENT: Table 5: please clarify that this is only for the 40 pigs subjected to both weighing, and the tape—the model estimation results, is it for the 40 (not for all pigs)? RESPONSE: Description on how the 40 pigs were obtained and used has been added in the text COMMENT: Submitted filename: RESPONSE TO REVIEWERS.docx Click here for additional data file. 22 Nov 2023 Heart girth best predicts live weights of market-age pigs in Tanzania PONE-D-23-16971R1 Dear Dr. Kabululu, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Anselme Shyaka, Ph.D Academic Editor PLOS ONE Additional Editor Comments (optional): Dear Dr. Kabululu, I am delighted to inform you that your paper has been accepted for publication by PLOS ONE. Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation. Reviewer #1: All comments have been addressed Reviewer #2: All comments have been addressed ********** 2. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes ********** 3. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: Yes ********** 4. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes ********** 5. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes ********** 6. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: The author has sufficiently - but really only just sufficiently - made revisions to the manuscript that address my comments. Some new papers have been referenced, but there is still a lack of critical comparison between this and the several earlier attempts to do something similar by other authors. As a real contribution to novelty, therefore, this paper remains lacking and in another journal it would probably be rejected again. PLoS One, however, has a policy of only looking at the technical sound-ness of the work, and not its more global context, so I am able to accept this, though somewhat reservedly. Reviewer #2: No additional comments. The research is relevant and adds more to evidence that is currently available. ********** 7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: No Reviewer #2: No ********** 24 Nov 2023 PONE-D-23-16971R1 Heart girth best predicts live weights of market-age pigs in Tanzania Dear Dr. Kabululu: I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org. If we can help with anything else, please email us at customercare@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Anselme Shyaka Academic Editor PLOS ONE